summaryrefslogtreecommitdiffstats
path: root/src/xz
diff options
context:
space:
mode:
Diffstat (limited to 'src/xz')
-rw-r--r--src/xz/Makefile.am133
-rw-r--r--src/xz/Makefile.in1162
-rw-r--r--src/xz/args.c763
-rw-r--r--src/xz/args.h44
-rw-r--r--src/xz/coder.c1109
-rw-r--r--src/xz/coder.h79
-rw-r--r--src/xz/file_io.c1403
-rw-r--r--src/xz/file_io.h179
-rw-r--r--src/xz/hardware.c338
-rw-r--r--src/xz/hardware.h75
-rw-r--r--src/xz/list.c1318
-rw-r--r--src/xz/list.h18
-rw-r--r--src/xz/main.c344
-rw-r--r--src/xz/main.h30
-rw-r--r--src/xz/message.c1146
-rw-r--r--src/xz/message.h156
-rw-r--r--src/xz/mytime.c86
-rw-r--r--src/xz/mytime.h43
-rw-r--r--src/xz/options.c359
-rw-r--r--src/xz/options.h31
-rw-r--r--src/xz/private.h66
-rw-r--r--src/xz/signals.c209
-rw-r--r--src/xz/signals.h43
-rw-r--r--src/xz/suffix.c401
-rw-r--r--src/xz/suffix.h36
-rw-r--r--src/xz/util.c287
-rw-r--r--src/xz/util.h119
-rw-r--r--src/xz/xz.13020
-rw-r--r--src/xz/xz_w32res.rc12
29 files changed, 13009 insertions, 0 deletions
diff --git a/src/xz/Makefile.am b/src/xz/Makefile.am
new file mode 100644
index 0000000..4bc64f3
--- /dev/null
+++ b/src/xz/Makefile.am
@@ -0,0 +1,133 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+bin_PROGRAMS = xz
+
+xz_SOURCES = \
+ args.c \
+ args.h \
+ coder.c \
+ coder.h \
+ file_io.c \
+ file_io.h \
+ hardware.c \
+ hardware.h \
+ main.c \
+ main.h \
+ message.c \
+ message.h \
+ mytime.c \
+ mytime.h \
+ options.c \
+ options.h \
+ private.h \
+ signals.c \
+ signals.h \
+ suffix.c \
+ suffix.h \
+ util.c \
+ util.h \
+ ../common/tuklib_open_stdxxx.c \
+ ../common/tuklib_progname.c \
+ ../common/tuklib_exit.c \
+ ../common/tuklib_mbstr_width.c \
+ ../common/tuklib_mbstr_fw.c
+
+if COND_MAIN_DECODER
+xz_SOURCES += \
+ list.c \
+ list.h
+endif
+
+if COND_W32
+xz_SOURCES += xz_w32res.rc
+endif
+
+xz_CPPFLAGS = \
+ -DLOCALEDIR=\"$(localedir)\" \
+ -I$(top_srcdir)/src/common \
+ -I$(top_srcdir)/src/liblzma/api \
+ -I$(top_builddir)/lib
+
+xz_LDADD = $(top_builddir)/src/liblzma/liblzma.la $(CAPSICUM_LIB)
+
+if COND_GNULIB
+xz_LDADD += $(top_builddir)/lib/libgnu.a
+endif
+
+# libgnu.a may need these libs, so this must be after libgnu.a.
+xz_LDADD += $(LTLIBINTL)
+
+
+# Windows resource compiler support
+.rc.o:
+ $(RC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) \
+ $(xz_CPPFLAGS) $(CPPFLAGS) $(RCFLAGS) -i $< -o $@
+
+
+dist_man_MANS = xz.1
+
+
+## Create symlinks for unxz and xzcat for convenience. Create symlinks also
+## for lzma, unlzma, and lzcat for compatibility with LZMA Utils 4.32.x.
+xzlinks = unxz xzcat
+
+if COND_LZMALINKS
+xzlinks += lzma unlzma lzcat
+endif
+
+install-exec-hook:
+ cd "$(DESTDIR)$(bindir)" && \
+ target=`echo xz | sed '$(transform)'`$(EXEEXT) && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'`$(LN_EXEEXT) && \
+ rm -f "$$link" && \
+ $(LN_S) "$$target" "$$link"; \
+ done
+
+# The installation of translated man pages abuses Automake internals
+# by calling "install-man" with redefined dist_man_MANS and man_MANS.
+# If this breaks some day, don't blame Automake developers.
+install-data-hook:
+ languages= ; \
+ if test "$(USE_NLS)" = yes && test -d "$(top_srcdir)/po4a/man"; then \
+ languages=`ls "$(top_srcdir)/po4a/man"`; \
+ fi; \
+ target=`echo xz | sed '$(transform)'` && \
+ for lang in . $$languages; do \
+ man="$(top_srcdir)/po4a/man/$$lang/xz.1" ; \
+ if test -f "$$man"; then \
+ $(MAKE) dist_man_MANS="$$man" man_MANS= \
+ mandir="$(mandir)/$$lang" install-man; \
+ fi; \
+ man1dir="$(DESTDIR)$(mandir)/$$lang/man1" && \
+ if test -f "$$man1dir/$$target.1"; then ( \
+ cd "$$man1dir" && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f "$$link.1" && \
+ $(LN_S) "$$target.1" "$$link.1"; \
+ done \
+ ); fi; \
+ done
+
+uninstall-hook:
+ cd "$(DESTDIR)$(bindir)" && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'`$(LN_EXEEXT) && \
+ rm -f "$$link"; \
+ done
+ languages= ; \
+ if test "$(USE_NLS)" = yes && test -d "$(top_srcdir)/po4a/man"; then \
+ languages=`ls "$(top_srcdir)/po4a/man"`; \
+ fi; \
+ for lang in . $$languages; do \
+ for name in xz $(xzlinks); do \
+ name=`echo $$name | sed '$(transform)'` && \
+ rm -f "$(DESTDIR)$(mandir)/$$lang/man1/$$name.1"; \
+ done; \
+ done
diff --git a/src/xz/Makefile.in b/src/xz/Makefile.in
new file mode 100644
index 0000000..e6422c5
--- /dev/null
+++ b/src/xz/Makefile.in
@@ -0,0 +1,1162 @@
+# Makefile.in generated by automake 1.16.5 from Makefile.am.
+# @configure_input@
+
+# Copyright (C) 1994-2021 Free Software Foundation, Inc.
+
+# This Makefile.in is free software; the Free Software Foundation
+# gives unlimited permission to copy and/or distribute it,
+# with or without modifications, as long as this notice is preserved.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY, to the extent permitted by law; without
+# even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+# PARTICULAR PURPOSE.
+
+@SET_MAKE@
+
+VPATH = @srcdir@
+am__is_gnu_make = { \
+ if test -z '$(MAKELEVEL)'; then \
+ false; \
+ elif test -n '$(MAKE_HOST)'; then \
+ true; \
+ elif test -n '$(MAKE_VERSION)' && test -n '$(CURDIR)'; then \
+ true; \
+ else \
+ false; \
+ fi; \
+}
+am__make_running_with_option = \
+ case $${target_option-} in \
+ ?) ;; \
+ *) echo "am__make_running_with_option: internal error: invalid" \
+ "target option '$${target_option-}' specified" >&2; \
+ exit 1;; \
+ esac; \
+ has_opt=no; \
+ sane_makeflags=$$MAKEFLAGS; \
+ if $(am__is_gnu_make); then \
+ sane_makeflags=$$MFLAGS; \
+ else \
+ case $$MAKEFLAGS in \
+ *\\[\ \ ]*) \
+ bs=\\; \
+ sane_makeflags=`printf '%s\n' "$$MAKEFLAGS" \
+ | sed "s/$$bs$$bs[$$bs $$bs ]*//g"`;; \
+ esac; \
+ fi; \
+ skip_next=no; \
+ strip_trailopt () \
+ { \
+ flg=`printf '%s\n' "$$flg" | sed "s/$$1.*$$//"`; \
+ }; \
+ for flg in $$sane_makeflags; do \
+ test $$skip_next = yes && { skip_next=no; continue; }; \
+ case $$flg in \
+ *=*|--*) continue;; \
+ -*I) strip_trailopt 'I'; skip_next=yes;; \
+ -*I?*) strip_trailopt 'I';; \
+ -*O) strip_trailopt 'O'; skip_next=yes;; \
+ -*O?*) strip_trailopt 'O';; \
+ -*l) strip_trailopt 'l'; skip_next=yes;; \
+ -*l?*) strip_trailopt 'l';; \
+ -[dEDm]) skip_next=yes;; \
+ -[JT]) skip_next=yes;; \
+ esac; \
+ case $$flg in \
+ *$$target_option*) has_opt=yes; break;; \
+ esac; \
+ done; \
+ test $$has_opt = yes
+am__make_dryrun = (target_option=n; $(am__make_running_with_option))
+am__make_keepgoing = (target_option=k; $(am__make_running_with_option))
+pkgdatadir = $(datadir)/@PACKAGE@
+pkgincludedir = $(includedir)/@PACKAGE@
+pkglibdir = $(libdir)/@PACKAGE@
+pkglibexecdir = $(libexecdir)/@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@
+bin_PROGRAMS = xz$(EXEEXT)
+@COND_MAIN_DECODER_TRUE@am__append_1 = \
+@COND_MAIN_DECODER_TRUE@ list.c \
+@COND_MAIN_DECODER_TRUE@ list.h
+
+@COND_W32_TRUE@am__append_2 = xz_w32res.rc
+@COND_GNULIB_TRUE@am__append_3 = $(top_builddir)/lib/libgnu.a
+@COND_LZMALINKS_TRUE@am__append_4 = lzma unlzma lzcat
+subdir = src/xz
+ACLOCAL_M4 = $(top_srcdir)/aclocal.m4
+am__aclocal_m4_deps = $(top_srcdir)/m4/ax_check_capsicum.m4 \
+ $(top_srcdir)/m4/ax_pthread.m4 $(top_srcdir)/m4/getopt.m4 \
+ $(top_srcdir)/m4/gettext.m4 $(top_srcdir)/m4/host-cpu-c-abi.m4 \
+ $(top_srcdir)/m4/iconv.m4 $(top_srcdir)/m4/intlmacosx.m4 \
+ $(top_srcdir)/m4/lib-ld.m4 $(top_srcdir)/m4/lib-link.m4 \
+ $(top_srcdir)/m4/lib-prefix.m4 $(top_srcdir)/m4/libtool.m4 \
+ $(top_srcdir)/m4/ltoptions.m4 $(top_srcdir)/m4/ltsugar.m4 \
+ $(top_srcdir)/m4/ltversion.m4 $(top_srcdir)/m4/lt~obsolete.m4 \
+ $(top_srcdir)/m4/nls.m4 $(top_srcdir)/m4/po.m4 \
+ $(top_srcdir)/m4/posix-shell.m4 $(top_srcdir)/m4/progtest.m4 \
+ $(top_srcdir)/m4/tuklib_common.m4 \
+ $(top_srcdir)/m4/tuklib_cpucores.m4 \
+ $(top_srcdir)/m4/tuklib_integer.m4 \
+ $(top_srcdir)/m4/tuklib_mbstr.m4 \
+ $(top_srcdir)/m4/tuklib_physmem.m4 \
+ $(top_srcdir)/m4/tuklib_progname.m4 \
+ $(top_srcdir)/m4/visibility.m4 $(top_srcdir)/configure.ac
+am__configure_deps = $(am__aclocal_m4_deps) $(CONFIGURE_DEPENDENCIES) \
+ $(ACLOCAL_M4)
+DIST_COMMON = $(srcdir)/Makefile.am $(am__DIST_COMMON)
+mkinstalldirs = $(install_sh) -d
+CONFIG_HEADER = $(top_builddir)/config.h
+CONFIG_CLEAN_FILES =
+CONFIG_CLEAN_VPATH_FILES =
+am__installdirs = "$(DESTDIR)$(bindir)" "$(DESTDIR)$(man1dir)"
+PROGRAMS = $(bin_PROGRAMS)
+am__xz_SOURCES_DIST = args.c args.h coder.c coder.h file_io.c \
+ file_io.h hardware.c hardware.h main.c main.h message.c \
+ message.h mytime.c mytime.h options.c options.h private.h \
+ signals.c signals.h suffix.c suffix.h util.c util.h \
+ ../common/tuklib_open_stdxxx.c ../common/tuklib_progname.c \
+ ../common/tuklib_exit.c ../common/tuklib_mbstr_width.c \
+ ../common/tuklib_mbstr_fw.c list.c list.h xz_w32res.rc
+@COND_MAIN_DECODER_TRUE@am__objects_1 = xz-list.$(OBJEXT)
+@COND_W32_TRUE@am__objects_2 = xz_w32res.$(OBJEXT)
+am_xz_OBJECTS = xz-args.$(OBJEXT) xz-coder.$(OBJEXT) \
+ xz-file_io.$(OBJEXT) xz-hardware.$(OBJEXT) xz-main.$(OBJEXT) \
+ xz-message.$(OBJEXT) xz-mytime.$(OBJEXT) xz-options.$(OBJEXT) \
+ xz-signals.$(OBJEXT) xz-suffix.$(OBJEXT) xz-util.$(OBJEXT) \
+ xz-tuklib_open_stdxxx.$(OBJEXT) xz-tuklib_progname.$(OBJEXT) \
+ xz-tuklib_exit.$(OBJEXT) xz-tuklib_mbstr_width.$(OBJEXT) \
+ xz-tuklib_mbstr_fw.$(OBJEXT) $(am__objects_1) $(am__objects_2)
+xz_OBJECTS = $(am_xz_OBJECTS)
+am__DEPENDENCIES_1 =
+xz_DEPENDENCIES = $(top_builddir)/src/liblzma/liblzma.la \
+ $(am__DEPENDENCIES_1) $(am__append_3) $(am__DEPENDENCIES_1)
+AM_V_lt = $(am__v_lt_@AM_V@)
+am__v_lt_ = $(am__v_lt_@AM_DEFAULT_V@)
+am__v_lt_0 = --silent
+am__v_lt_1 =
+AM_V_P = $(am__v_P_@AM_V@)
+am__v_P_ = $(am__v_P_@AM_DEFAULT_V@)
+am__v_P_0 = false
+am__v_P_1 = :
+AM_V_GEN = $(am__v_GEN_@AM_V@)
+am__v_GEN_ = $(am__v_GEN_@AM_DEFAULT_V@)
+am__v_GEN_0 = @echo " GEN " $@;
+am__v_GEN_1 =
+AM_V_at = $(am__v_at_@AM_V@)
+am__v_at_ = $(am__v_at_@AM_DEFAULT_V@)
+am__v_at_0 = @
+am__v_at_1 =
+DEFAULT_INCLUDES = -I.@am__isrc@ -I$(top_builddir)
+depcomp = $(SHELL) $(top_srcdir)/build-aux/depcomp
+am__maybe_remake_depfiles = depfiles
+am__depfiles_remade = ./$(DEPDIR)/xz-args.Po ./$(DEPDIR)/xz-coder.Po \
+ ./$(DEPDIR)/xz-file_io.Po ./$(DEPDIR)/xz-hardware.Po \
+ ./$(DEPDIR)/xz-list.Po ./$(DEPDIR)/xz-main.Po \
+ ./$(DEPDIR)/xz-message.Po ./$(DEPDIR)/xz-mytime.Po \
+ ./$(DEPDIR)/xz-options.Po ./$(DEPDIR)/xz-signals.Po \
+ ./$(DEPDIR)/xz-suffix.Po ./$(DEPDIR)/xz-tuklib_exit.Po \
+ ./$(DEPDIR)/xz-tuklib_mbstr_fw.Po \
+ ./$(DEPDIR)/xz-tuklib_mbstr_width.Po \
+ ./$(DEPDIR)/xz-tuklib_open_stdxxx.Po \
+ ./$(DEPDIR)/xz-tuklib_progname.Po ./$(DEPDIR)/xz-util.Po
+am__mv = mv -f
+COMPILE = $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) \
+ $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS)
+LTCOMPILE = $(LIBTOOL) $(AM_V_lt) --tag=CC $(AM_LIBTOOLFLAGS) \
+ $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) \
+ $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) \
+ $(AM_CFLAGS) $(CFLAGS)
+AM_V_CC = $(am__v_CC_@AM_V@)
+am__v_CC_ = $(am__v_CC_@AM_DEFAULT_V@)
+am__v_CC_0 = @echo " CC " $@;
+am__v_CC_1 =
+CCLD = $(CC)
+LINK = $(LIBTOOL) $(AM_V_lt) --tag=CC $(AM_LIBTOOLFLAGS) \
+ $(LIBTOOLFLAGS) --mode=link $(CCLD) $(AM_CFLAGS) $(CFLAGS) \
+ $(AM_LDFLAGS) $(LDFLAGS) -o $@
+AM_V_CCLD = $(am__v_CCLD_@AM_V@)
+am__v_CCLD_ = $(am__v_CCLD_@AM_DEFAULT_V@)
+am__v_CCLD_0 = @echo " CCLD " $@;
+am__v_CCLD_1 =
+SOURCES = $(xz_SOURCES)
+DIST_SOURCES = $(am__xz_SOURCES_DIST)
+am__can_run_installinfo = \
+ case $$AM_UPDATE_INFO_DIR in \
+ n|no|NO) false;; \
+ *) (install-info --version) >/dev/null 2>&1;; \
+ esac
+am__vpath_adj_setup = srcdirstrip=`echo "$(srcdir)" | sed 's|.|.|g'`;
+am__vpath_adj = case $$p in \
+ $(srcdir)/*) f=`echo "$$p" | sed "s|^$$srcdirstrip/||"`;; \
+ *) f=$$p;; \
+ esac;
+am__strip_dir = f=`echo $$p | sed -e 's|^.*/||'`;
+am__install_max = 40
+am__nobase_strip_setup = \
+ srcdirstrip=`echo "$(srcdir)" | sed 's/[].[^$$\\*|]/\\\\&/g'`
+am__nobase_strip = \
+ for p in $$list; do echo "$$p"; done | sed -e "s|$$srcdirstrip/||"
+am__nobase_list = $(am__nobase_strip_setup); \
+ for p in $$list; do echo "$$p $$p"; done | \
+ sed "s| $$srcdirstrip/| |;"' / .*\//!s/ .*/ ./; s,\( .*\)/[^/]*$$,\1,' | \
+ $(AWK) 'BEGIN { files["."] = "" } { files[$$2] = files[$$2] " " $$1; \
+ if (++n[$$2] == $(am__install_max)) \
+ { print $$2, files[$$2]; n[$$2] = 0; files[$$2] = "" } } \
+ END { for (dir in files) print dir, files[dir] }'
+am__base_list = \
+ sed '$$!N;$$!N;$$!N;$$!N;$$!N;$$!N;$$!N;s/\n/ /g' | \
+ sed '$$!N;$$!N;$$!N;$$!N;s/\n/ /g'
+am__uninstall_files_from_dir = { \
+ test -z "$$files" \
+ || { test ! -d "$$dir" && test ! -f "$$dir" && test ! -r "$$dir"; } \
+ || { echo " ( cd '$$dir' && rm -f" $$files ")"; \
+ $(am__cd) "$$dir" && rm -f $$files; }; \
+ }
+man1dir = $(mandir)/man1
+NROFF = nroff
+MANS = $(dist_man_MANS)
+am__tagged_files = $(HEADERS) $(SOURCES) $(TAGS_FILES) $(LISP)
+# Read a list of newline-separated strings from the standard input,
+# and print each of them once, without duplicates. Input order is
+# *not* preserved.
+am__uniquify_input = $(AWK) '\
+ BEGIN { nonempty = 0; } \
+ { items[$$0] = 1; nonempty = 1; } \
+ END { if (nonempty) { for (i in items) print i; }; } \
+'
+# Make sure the list of sources is unique. This is necessary because,
+# e.g., the same source file might be shared among _SOURCES variables
+# for different programs/libraries.
+am__define_uniq_tagged_files = \
+ list='$(am__tagged_files)'; \
+ unique=`for i in $$list; do \
+ if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \
+ done | $(am__uniquify_input)`
+am__DIST_COMMON = $(dist_man_MANS) $(srcdir)/Makefile.in \
+ $(top_srcdir)/build-aux/depcomp
+DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
+ACLOCAL = @ACLOCAL@
+AMTAR = @AMTAR@
+AM_CFLAGS = @AM_CFLAGS@
+AM_DEFAULT_VERBOSITY = @AM_DEFAULT_VERBOSITY@
+AR = @AR@
+AS = @AS@
+AUTOCONF = @AUTOCONF@
+AUTOHEADER = @AUTOHEADER@
+AUTOMAKE = @AUTOMAKE@
+AWK = @AWK@
+CAPSICUM_LIB = @CAPSICUM_LIB@
+CC = @CC@
+CCAS = @CCAS@
+CCASDEPMODE = @CCASDEPMODE@
+CCASFLAGS = @CCASFLAGS@
+CCDEPMODE = @CCDEPMODE@
+CFLAGS = @CFLAGS@
+CFLAG_VISIBILITY = @CFLAG_VISIBILITY@
+CPP = @CPP@
+CPPFLAGS = @CPPFLAGS@
+CSCOPE = @CSCOPE@
+CTAGS = @CTAGS@
+CYGPATH_W = @CYGPATH_W@
+DEFS = @DEFS@
+DEPDIR = @DEPDIR@
+DLLTOOL = @DLLTOOL@
+DSYMUTIL = @DSYMUTIL@
+DUMPBIN = @DUMPBIN@
+ECHO_C = @ECHO_C@
+ECHO_N = @ECHO_N@
+ECHO_T = @ECHO_T@
+EGREP = @EGREP@
+ETAGS = @ETAGS@
+EXEEXT = @EXEEXT@
+FGREP = @FGREP@
+FILECMD = @FILECMD@
+GETOPT_H = @GETOPT_H@
+GETTEXT_MACRO_VERSION = @GETTEXT_MACRO_VERSION@
+GMSGFMT = @GMSGFMT@
+GMSGFMT_015 = @GMSGFMT_015@
+GREP = @GREP@
+HAVE_VISIBILITY = @HAVE_VISIBILITY@
+INSTALL = @INSTALL@
+INSTALL_DATA = @INSTALL_DATA@
+INSTALL_PROGRAM = @INSTALL_PROGRAM@
+INSTALL_SCRIPT = @INSTALL_SCRIPT@
+INSTALL_STRIP_PROGRAM = @INSTALL_STRIP_PROGRAM@
+INTLLIBS = @INTLLIBS@
+INTL_MACOSX_LIBS = @INTL_MACOSX_LIBS@
+LD = @LD@
+LDFLAGS = @LDFLAGS@
+LIBICONV = @LIBICONV@
+LIBINTL = @LIBINTL@
+LIBOBJS = @LIBOBJS@
+LIBS = @LIBS@
+LIBTOOL = @LIBTOOL@
+LIPO = @LIPO@
+LN_EXEEXT = @LN_EXEEXT@
+LN_S = @LN_S@
+LTLIBICONV = @LTLIBICONV@
+LTLIBINTL = @LTLIBINTL@
+LTLIBOBJS = @LTLIBOBJS@
+LT_SYS_LIBRARY_PATH = @LT_SYS_LIBRARY_PATH@
+MAKEINFO = @MAKEINFO@
+MANIFEST_TOOL = @MANIFEST_TOOL@
+MKDIR_P = @MKDIR_P@
+MSGFMT = @MSGFMT@
+MSGMERGE = @MSGMERGE@
+MSGMERGE_FOR_MSGFMT_OPTION = @MSGMERGE_FOR_MSGFMT_OPTION@
+NM = @NM@
+NMEDIT = @NMEDIT@
+OBJDUMP = @OBJDUMP@
+OBJEXT = @OBJEXT@
+OTOOL = @OTOOL@
+OTOOL64 = @OTOOL64@
+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@
+POSIX_SHELL = @POSIX_SHELL@
+POSUB = @POSUB@
+PREFERABLY_POSIX_SHELL = @PREFERABLY_POSIX_SHELL@
+PTHREAD_CC = @PTHREAD_CC@
+PTHREAD_CFLAGS = @PTHREAD_CFLAGS@
+PTHREAD_CXX = @PTHREAD_CXX@
+PTHREAD_LIBS = @PTHREAD_LIBS@
+RANLIB = @RANLIB@
+RC = @RC@
+SED = @SED@
+SET_MAKE = @SET_MAKE@
+SHELL = @SHELL@
+STRIP = @STRIP@
+USE_NLS = @USE_NLS@
+VERSION = @VERSION@
+XGETTEXT = @XGETTEXT@
+XGETTEXT_015 = @XGETTEXT_015@
+XGETTEXT_EXTRA_OPTIONS = @XGETTEXT_EXTRA_OPTIONS@
+abs_builddir = @abs_builddir@
+abs_srcdir = @abs_srcdir@
+abs_top_builddir = @abs_top_builddir@
+abs_top_srcdir = @abs_top_srcdir@
+ac_ct_AR = @ac_ct_AR@
+ac_ct_CC = @ac_ct_CC@
+ac_ct_DUMPBIN = @ac_ct_DUMPBIN@
+am__include = @am__include@
+am__leading_dot = @am__leading_dot@
+am__quote = @am__quote@
+am__tar = @am__tar@
+am__untar = @am__untar@
+ax_pthread_config = @ax_pthread_config@
+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@
+enable_path_for_scripts = @enable_path_for_scripts@
+exec_prefix = @exec_prefix@
+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@
+localedir = @localedir@
+localstatedir = @localstatedir@
+mandir = @mandir@
+mkdir_p = @mkdir_p@
+oldincludedir = @oldincludedir@
+pdfdir = @pdfdir@
+prefix = @prefix@
+program_transform_name = @program_transform_name@
+psdir = @psdir@
+runstatedir = @runstatedir@
+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@
+xz = @xz@
+xz_SOURCES = args.c args.h coder.c coder.h file_io.c file_io.h \
+ hardware.c hardware.h main.c main.h message.c message.h \
+ mytime.c mytime.h options.c options.h private.h signals.c \
+ signals.h suffix.c suffix.h util.c util.h \
+ ../common/tuklib_open_stdxxx.c ../common/tuklib_progname.c \
+ ../common/tuklib_exit.c ../common/tuklib_mbstr_width.c \
+ ../common/tuklib_mbstr_fw.c $(am__append_1) $(am__append_2)
+xz_CPPFLAGS = \
+ -DLOCALEDIR=\"$(localedir)\" \
+ -I$(top_srcdir)/src/common \
+ -I$(top_srcdir)/src/liblzma/api \
+ -I$(top_builddir)/lib
+
+
+# libgnu.a may need these libs, so this must be after libgnu.a.
+xz_LDADD = $(top_builddir)/src/liblzma/liblzma.la $(CAPSICUM_LIB) \
+ $(am__append_3) $(LTLIBINTL)
+dist_man_MANS = xz.1
+xzlinks = unxz xzcat $(am__append_4)
+all: all-am
+
+.SUFFIXES:
+.SUFFIXES: .c .lo .o .obj .rc
+$(srcdir)/Makefile.in: $(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) --foreign src/xz/Makefile'; \
+ $(am__cd) $(top_srcdir) && \
+ $(AUTOMAKE) --foreign src/xz/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__maybe_remake_depfiles)'; \
+ cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@ $(am__maybe_remake_depfiles);; \
+ esac;
+
+$(top_builddir)/config.status: $(top_srcdir)/configure $(CONFIG_STATUS_DEPENDENCIES)
+ cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh
+
+$(top_srcdir)/configure: $(am__configure_deps)
+ cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh
+$(ACLOCAL_M4): $(am__aclocal_m4_deps)
+ cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh
+$(am__aclocal_m4_deps):
+install-binPROGRAMS: $(bin_PROGRAMS)
+ @$(NORMAL_INSTALL)
+ @list='$(bin_PROGRAMS)'; test -n "$(bindir)" || list=; \
+ if test -n "$$list"; then \
+ echo " $(MKDIR_P) '$(DESTDIR)$(bindir)'"; \
+ $(MKDIR_P) "$(DESTDIR)$(bindir)" || exit 1; \
+ fi; \
+ for p in $$list; do echo "$$p $$p"; done | \
+ sed 's/$(EXEEXT)$$//' | \
+ while read p p1; do if test -f $$p \
+ || test -f $$p1 \
+ ; then echo "$$p"; echo "$$p"; else :; fi; \
+ done | \
+ sed -e 'p;s,.*/,,;n;h' \
+ -e 's|.*|.|' \
+ -e 'p;x;s,.*/,,;s/$(EXEEXT)$$//;$(transform);s/$$/$(EXEEXT)/' | \
+ sed 'N;N;N;s,\n, ,g' | \
+ $(AWK) 'BEGIN { files["."] = ""; dirs["."] = 1 } \
+ { d=$$3; if (dirs[d] != 1) { print "d", d; dirs[d] = 1 } \
+ if ($$2 == $$4) files[d] = files[d] " " $$1; \
+ else { print "f", $$3 "/" $$4, $$1; } } \
+ END { for (d in files) print "f", d, files[d] }' | \
+ while read type dir files; do \
+ if test "$$dir" = .; then dir=; else dir=/$$dir; fi; \
+ test -z "$$files" || { \
+ echo " $(INSTALL_PROGRAM_ENV) $(LIBTOOL) $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=install $(INSTALL_PROGRAM) $$files '$(DESTDIR)$(bindir)$$dir'"; \
+ $(INSTALL_PROGRAM_ENV) $(LIBTOOL) $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=install $(INSTALL_PROGRAM) $$files "$(DESTDIR)$(bindir)$$dir" || exit $$?; \
+ } \
+ ; done
+
+uninstall-binPROGRAMS:
+ @$(NORMAL_UNINSTALL)
+ @list='$(bin_PROGRAMS)'; test -n "$(bindir)" || list=; \
+ files=`for p in $$list; do echo "$$p"; done | \
+ sed -e 'h;s,^.*/,,;s/$(EXEEXT)$$//;$(transform)' \
+ -e 's/$$/$(EXEEXT)/' \
+ `; \
+ test -n "$$list" || exit 0; \
+ echo " ( cd '$(DESTDIR)$(bindir)' && rm -f" $$files ")"; \
+ cd "$(DESTDIR)$(bindir)" && rm -f $$files
+
+clean-binPROGRAMS:
+ @list='$(bin_PROGRAMS)'; test -n "$$list" || exit 0; \
+ echo " rm -f" $$list; \
+ rm -f $$list || exit $$?; \
+ test -n "$(EXEEXT)" || exit 0; \
+ list=`for p in $$list; do echo "$$p"; done | sed 's/$(EXEEXT)$$//'`; \
+ echo " rm -f" $$list; \
+ rm -f $$list
+
+xz$(EXEEXT): $(xz_OBJECTS) $(xz_DEPENDENCIES) $(EXTRA_xz_DEPENDENCIES)
+ @rm -f xz$(EXEEXT)
+ $(AM_V_CCLD)$(LINK) $(xz_OBJECTS) $(xz_LDADD) $(LIBS)
+
+mostlyclean-compile:
+ -rm -f *.$(OBJEXT)
+
+distclean-compile:
+ -rm -f *.tab.c
+
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-args.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-coder.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-file_io.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-hardware.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-list.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-main.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-message.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-mytime.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-options.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-signals.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-suffix.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-tuklib_exit.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-tuklib_mbstr_fw.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-tuklib_mbstr_width.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-tuklib_open_stdxxx.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-tuklib_progname.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xz-util.Po@am__quote@ # am--include-marker
+
+$(am__depfiles_remade):
+ @$(MKDIR_P) $(@D)
+ @echo '# dummy' >$@-t && $(am__mv) $@-t $@
+
+am--depfiles: $(am__depfiles_remade)
+
+.c.o:
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(COMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo -c -o $@ $<
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='$<' object='$@' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(COMPILE) -c -o $@ $<
+
+.c.obj:
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(COMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo -c -o $@ `$(CYGPATH_W) '$<'`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='$<' object='$@' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(COMPILE) -c -o $@ `$(CYGPATH_W) '$<'`
+
+.c.lo:
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(LTCOMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tpo -c -o $@ $<
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/$*.Tpo $(DEPDIR)/$*.Plo
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='$<' object='$@' libtool=yes @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(LTCOMPILE) -c -o $@ $<
+
+xz-args.o: args.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-args.o -MD -MP -MF $(DEPDIR)/xz-args.Tpo -c -o xz-args.o `test -f 'args.c' || echo '$(srcdir)/'`args.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-args.Tpo $(DEPDIR)/xz-args.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='args.c' object='xz-args.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-args.o `test -f 'args.c' || echo '$(srcdir)/'`args.c
+
+xz-args.obj: args.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-args.obj -MD -MP -MF $(DEPDIR)/xz-args.Tpo -c -o xz-args.obj `if test -f 'args.c'; then $(CYGPATH_W) 'args.c'; else $(CYGPATH_W) '$(srcdir)/args.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-args.Tpo $(DEPDIR)/xz-args.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='args.c' object='xz-args.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-args.obj `if test -f 'args.c'; then $(CYGPATH_W) 'args.c'; else $(CYGPATH_W) '$(srcdir)/args.c'; fi`
+
+xz-coder.o: coder.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-coder.o -MD -MP -MF $(DEPDIR)/xz-coder.Tpo -c -o xz-coder.o `test -f 'coder.c' || echo '$(srcdir)/'`coder.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-coder.Tpo $(DEPDIR)/xz-coder.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='coder.c' object='xz-coder.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-coder.o `test -f 'coder.c' || echo '$(srcdir)/'`coder.c
+
+xz-coder.obj: coder.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-coder.obj -MD -MP -MF $(DEPDIR)/xz-coder.Tpo -c -o xz-coder.obj `if test -f 'coder.c'; then $(CYGPATH_W) 'coder.c'; else $(CYGPATH_W) '$(srcdir)/coder.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-coder.Tpo $(DEPDIR)/xz-coder.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='coder.c' object='xz-coder.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-coder.obj `if test -f 'coder.c'; then $(CYGPATH_W) 'coder.c'; else $(CYGPATH_W) '$(srcdir)/coder.c'; fi`
+
+xz-file_io.o: file_io.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-file_io.o -MD -MP -MF $(DEPDIR)/xz-file_io.Tpo -c -o xz-file_io.o `test -f 'file_io.c' || echo '$(srcdir)/'`file_io.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-file_io.Tpo $(DEPDIR)/xz-file_io.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='file_io.c' object='xz-file_io.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-file_io.o `test -f 'file_io.c' || echo '$(srcdir)/'`file_io.c
+
+xz-file_io.obj: file_io.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-file_io.obj -MD -MP -MF $(DEPDIR)/xz-file_io.Tpo -c -o xz-file_io.obj `if test -f 'file_io.c'; then $(CYGPATH_W) 'file_io.c'; else $(CYGPATH_W) '$(srcdir)/file_io.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-file_io.Tpo $(DEPDIR)/xz-file_io.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='file_io.c' object='xz-file_io.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-file_io.obj `if test -f 'file_io.c'; then $(CYGPATH_W) 'file_io.c'; else $(CYGPATH_W) '$(srcdir)/file_io.c'; fi`
+
+xz-hardware.o: hardware.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-hardware.o -MD -MP -MF $(DEPDIR)/xz-hardware.Tpo -c -o xz-hardware.o `test -f 'hardware.c' || echo '$(srcdir)/'`hardware.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-hardware.Tpo $(DEPDIR)/xz-hardware.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='hardware.c' object='xz-hardware.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-hardware.o `test -f 'hardware.c' || echo '$(srcdir)/'`hardware.c
+
+xz-hardware.obj: hardware.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-hardware.obj -MD -MP -MF $(DEPDIR)/xz-hardware.Tpo -c -o xz-hardware.obj `if test -f 'hardware.c'; then $(CYGPATH_W) 'hardware.c'; else $(CYGPATH_W) '$(srcdir)/hardware.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-hardware.Tpo $(DEPDIR)/xz-hardware.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='hardware.c' object='xz-hardware.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-hardware.obj `if test -f 'hardware.c'; then $(CYGPATH_W) 'hardware.c'; else $(CYGPATH_W) '$(srcdir)/hardware.c'; fi`
+
+xz-main.o: main.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-main.o -MD -MP -MF $(DEPDIR)/xz-main.Tpo -c -o xz-main.o `test -f 'main.c' || echo '$(srcdir)/'`main.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-main.Tpo $(DEPDIR)/xz-main.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='main.c' object='xz-main.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-main.o `test -f 'main.c' || echo '$(srcdir)/'`main.c
+
+xz-main.obj: main.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-main.obj -MD -MP -MF $(DEPDIR)/xz-main.Tpo -c -o xz-main.obj `if test -f 'main.c'; then $(CYGPATH_W) 'main.c'; else $(CYGPATH_W) '$(srcdir)/main.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-main.Tpo $(DEPDIR)/xz-main.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='main.c' object='xz-main.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-main.obj `if test -f 'main.c'; then $(CYGPATH_W) 'main.c'; else $(CYGPATH_W) '$(srcdir)/main.c'; fi`
+
+xz-message.o: message.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-message.o -MD -MP -MF $(DEPDIR)/xz-message.Tpo -c -o xz-message.o `test -f 'message.c' || echo '$(srcdir)/'`message.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-message.Tpo $(DEPDIR)/xz-message.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='message.c' object='xz-message.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-message.o `test -f 'message.c' || echo '$(srcdir)/'`message.c
+
+xz-message.obj: message.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-message.obj -MD -MP -MF $(DEPDIR)/xz-message.Tpo -c -o xz-message.obj `if test -f 'message.c'; then $(CYGPATH_W) 'message.c'; else $(CYGPATH_W) '$(srcdir)/message.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-message.Tpo $(DEPDIR)/xz-message.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='message.c' object='xz-message.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-message.obj `if test -f 'message.c'; then $(CYGPATH_W) 'message.c'; else $(CYGPATH_W) '$(srcdir)/message.c'; fi`
+
+xz-mytime.o: mytime.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-mytime.o -MD -MP -MF $(DEPDIR)/xz-mytime.Tpo -c -o xz-mytime.o `test -f 'mytime.c' || echo '$(srcdir)/'`mytime.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-mytime.Tpo $(DEPDIR)/xz-mytime.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='mytime.c' object='xz-mytime.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-mytime.o `test -f 'mytime.c' || echo '$(srcdir)/'`mytime.c
+
+xz-mytime.obj: mytime.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-mytime.obj -MD -MP -MF $(DEPDIR)/xz-mytime.Tpo -c -o xz-mytime.obj `if test -f 'mytime.c'; then $(CYGPATH_W) 'mytime.c'; else $(CYGPATH_W) '$(srcdir)/mytime.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-mytime.Tpo $(DEPDIR)/xz-mytime.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='mytime.c' object='xz-mytime.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-mytime.obj `if test -f 'mytime.c'; then $(CYGPATH_W) 'mytime.c'; else $(CYGPATH_W) '$(srcdir)/mytime.c'; fi`
+
+xz-options.o: options.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-options.o -MD -MP -MF $(DEPDIR)/xz-options.Tpo -c -o xz-options.o `test -f 'options.c' || echo '$(srcdir)/'`options.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-options.Tpo $(DEPDIR)/xz-options.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='options.c' object='xz-options.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-options.o `test -f 'options.c' || echo '$(srcdir)/'`options.c
+
+xz-options.obj: options.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-options.obj -MD -MP -MF $(DEPDIR)/xz-options.Tpo -c -o xz-options.obj `if test -f 'options.c'; then $(CYGPATH_W) 'options.c'; else $(CYGPATH_W) '$(srcdir)/options.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-options.Tpo $(DEPDIR)/xz-options.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='options.c' object='xz-options.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-options.obj `if test -f 'options.c'; then $(CYGPATH_W) 'options.c'; else $(CYGPATH_W) '$(srcdir)/options.c'; fi`
+
+xz-signals.o: signals.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-signals.o -MD -MP -MF $(DEPDIR)/xz-signals.Tpo -c -o xz-signals.o `test -f 'signals.c' || echo '$(srcdir)/'`signals.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-signals.Tpo $(DEPDIR)/xz-signals.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='signals.c' object='xz-signals.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-signals.o `test -f 'signals.c' || echo '$(srcdir)/'`signals.c
+
+xz-signals.obj: signals.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-signals.obj -MD -MP -MF $(DEPDIR)/xz-signals.Tpo -c -o xz-signals.obj `if test -f 'signals.c'; then $(CYGPATH_W) 'signals.c'; else $(CYGPATH_W) '$(srcdir)/signals.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-signals.Tpo $(DEPDIR)/xz-signals.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='signals.c' object='xz-signals.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-signals.obj `if test -f 'signals.c'; then $(CYGPATH_W) 'signals.c'; else $(CYGPATH_W) '$(srcdir)/signals.c'; fi`
+
+xz-suffix.o: suffix.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-suffix.o -MD -MP -MF $(DEPDIR)/xz-suffix.Tpo -c -o xz-suffix.o `test -f 'suffix.c' || echo '$(srcdir)/'`suffix.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-suffix.Tpo $(DEPDIR)/xz-suffix.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='suffix.c' object='xz-suffix.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-suffix.o `test -f 'suffix.c' || echo '$(srcdir)/'`suffix.c
+
+xz-suffix.obj: suffix.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-suffix.obj -MD -MP -MF $(DEPDIR)/xz-suffix.Tpo -c -o xz-suffix.obj `if test -f 'suffix.c'; then $(CYGPATH_W) 'suffix.c'; else $(CYGPATH_W) '$(srcdir)/suffix.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-suffix.Tpo $(DEPDIR)/xz-suffix.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='suffix.c' object='xz-suffix.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-suffix.obj `if test -f 'suffix.c'; then $(CYGPATH_W) 'suffix.c'; else $(CYGPATH_W) '$(srcdir)/suffix.c'; fi`
+
+xz-util.o: util.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-util.o -MD -MP -MF $(DEPDIR)/xz-util.Tpo -c -o xz-util.o `test -f 'util.c' || echo '$(srcdir)/'`util.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-util.Tpo $(DEPDIR)/xz-util.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='util.c' object='xz-util.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-util.o `test -f 'util.c' || echo '$(srcdir)/'`util.c
+
+xz-util.obj: util.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-util.obj -MD -MP -MF $(DEPDIR)/xz-util.Tpo -c -o xz-util.obj `if test -f 'util.c'; then $(CYGPATH_W) 'util.c'; else $(CYGPATH_W) '$(srcdir)/util.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-util.Tpo $(DEPDIR)/xz-util.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='util.c' object='xz-util.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-util.obj `if test -f 'util.c'; then $(CYGPATH_W) 'util.c'; else $(CYGPATH_W) '$(srcdir)/util.c'; fi`
+
+xz-tuklib_open_stdxxx.o: ../common/tuklib_open_stdxxx.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_open_stdxxx.o -MD -MP -MF $(DEPDIR)/xz-tuklib_open_stdxxx.Tpo -c -o xz-tuklib_open_stdxxx.o `test -f '../common/tuklib_open_stdxxx.c' || echo '$(srcdir)/'`../common/tuklib_open_stdxxx.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_open_stdxxx.Tpo $(DEPDIR)/xz-tuklib_open_stdxxx.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_open_stdxxx.c' object='xz-tuklib_open_stdxxx.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_open_stdxxx.o `test -f '../common/tuklib_open_stdxxx.c' || echo '$(srcdir)/'`../common/tuklib_open_stdxxx.c
+
+xz-tuklib_open_stdxxx.obj: ../common/tuklib_open_stdxxx.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_open_stdxxx.obj -MD -MP -MF $(DEPDIR)/xz-tuklib_open_stdxxx.Tpo -c -o xz-tuklib_open_stdxxx.obj `if test -f '../common/tuklib_open_stdxxx.c'; then $(CYGPATH_W) '../common/tuklib_open_stdxxx.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_open_stdxxx.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_open_stdxxx.Tpo $(DEPDIR)/xz-tuklib_open_stdxxx.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_open_stdxxx.c' object='xz-tuklib_open_stdxxx.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_open_stdxxx.obj `if test -f '../common/tuklib_open_stdxxx.c'; then $(CYGPATH_W) '../common/tuklib_open_stdxxx.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_open_stdxxx.c'; fi`
+
+xz-tuklib_progname.o: ../common/tuklib_progname.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_progname.o -MD -MP -MF $(DEPDIR)/xz-tuklib_progname.Tpo -c -o xz-tuklib_progname.o `test -f '../common/tuklib_progname.c' || echo '$(srcdir)/'`../common/tuklib_progname.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_progname.Tpo $(DEPDIR)/xz-tuklib_progname.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_progname.c' object='xz-tuklib_progname.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_progname.o `test -f '../common/tuklib_progname.c' || echo '$(srcdir)/'`../common/tuklib_progname.c
+
+xz-tuklib_progname.obj: ../common/tuklib_progname.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_progname.obj -MD -MP -MF $(DEPDIR)/xz-tuklib_progname.Tpo -c -o xz-tuklib_progname.obj `if test -f '../common/tuklib_progname.c'; then $(CYGPATH_W) '../common/tuklib_progname.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_progname.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_progname.Tpo $(DEPDIR)/xz-tuklib_progname.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_progname.c' object='xz-tuklib_progname.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_progname.obj `if test -f '../common/tuklib_progname.c'; then $(CYGPATH_W) '../common/tuklib_progname.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_progname.c'; fi`
+
+xz-tuklib_exit.o: ../common/tuklib_exit.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_exit.o -MD -MP -MF $(DEPDIR)/xz-tuklib_exit.Tpo -c -o xz-tuklib_exit.o `test -f '../common/tuklib_exit.c' || echo '$(srcdir)/'`../common/tuklib_exit.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_exit.Tpo $(DEPDIR)/xz-tuklib_exit.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_exit.c' object='xz-tuklib_exit.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_exit.o `test -f '../common/tuklib_exit.c' || echo '$(srcdir)/'`../common/tuklib_exit.c
+
+xz-tuklib_exit.obj: ../common/tuklib_exit.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_exit.obj -MD -MP -MF $(DEPDIR)/xz-tuklib_exit.Tpo -c -o xz-tuklib_exit.obj `if test -f '../common/tuklib_exit.c'; then $(CYGPATH_W) '../common/tuklib_exit.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_exit.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_exit.Tpo $(DEPDIR)/xz-tuklib_exit.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_exit.c' object='xz-tuklib_exit.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_exit.obj `if test -f '../common/tuklib_exit.c'; then $(CYGPATH_W) '../common/tuklib_exit.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_exit.c'; fi`
+
+xz-tuklib_mbstr_width.o: ../common/tuklib_mbstr_width.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_mbstr_width.o -MD -MP -MF $(DEPDIR)/xz-tuklib_mbstr_width.Tpo -c -o xz-tuklib_mbstr_width.o `test -f '../common/tuklib_mbstr_width.c' || echo '$(srcdir)/'`../common/tuklib_mbstr_width.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_mbstr_width.Tpo $(DEPDIR)/xz-tuklib_mbstr_width.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_mbstr_width.c' object='xz-tuklib_mbstr_width.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_mbstr_width.o `test -f '../common/tuklib_mbstr_width.c' || echo '$(srcdir)/'`../common/tuklib_mbstr_width.c
+
+xz-tuklib_mbstr_width.obj: ../common/tuklib_mbstr_width.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_mbstr_width.obj -MD -MP -MF $(DEPDIR)/xz-tuklib_mbstr_width.Tpo -c -o xz-tuklib_mbstr_width.obj `if test -f '../common/tuklib_mbstr_width.c'; then $(CYGPATH_W) '../common/tuklib_mbstr_width.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_mbstr_width.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_mbstr_width.Tpo $(DEPDIR)/xz-tuklib_mbstr_width.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_mbstr_width.c' object='xz-tuklib_mbstr_width.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_mbstr_width.obj `if test -f '../common/tuklib_mbstr_width.c'; then $(CYGPATH_W) '../common/tuklib_mbstr_width.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_mbstr_width.c'; fi`
+
+xz-tuklib_mbstr_fw.o: ../common/tuklib_mbstr_fw.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_mbstr_fw.o -MD -MP -MF $(DEPDIR)/xz-tuklib_mbstr_fw.Tpo -c -o xz-tuklib_mbstr_fw.o `test -f '../common/tuklib_mbstr_fw.c' || echo '$(srcdir)/'`../common/tuklib_mbstr_fw.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_mbstr_fw.Tpo $(DEPDIR)/xz-tuklib_mbstr_fw.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_mbstr_fw.c' object='xz-tuklib_mbstr_fw.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_mbstr_fw.o `test -f '../common/tuklib_mbstr_fw.c' || echo '$(srcdir)/'`../common/tuklib_mbstr_fw.c
+
+xz-tuklib_mbstr_fw.obj: ../common/tuklib_mbstr_fw.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-tuklib_mbstr_fw.obj -MD -MP -MF $(DEPDIR)/xz-tuklib_mbstr_fw.Tpo -c -o xz-tuklib_mbstr_fw.obj `if test -f '../common/tuklib_mbstr_fw.c'; then $(CYGPATH_W) '../common/tuklib_mbstr_fw.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_mbstr_fw.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-tuklib_mbstr_fw.Tpo $(DEPDIR)/xz-tuklib_mbstr_fw.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='../common/tuklib_mbstr_fw.c' object='xz-tuklib_mbstr_fw.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-tuklib_mbstr_fw.obj `if test -f '../common/tuklib_mbstr_fw.c'; then $(CYGPATH_W) '../common/tuklib_mbstr_fw.c'; else $(CYGPATH_W) '$(srcdir)/../common/tuklib_mbstr_fw.c'; fi`
+
+xz-list.o: list.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-list.o -MD -MP -MF $(DEPDIR)/xz-list.Tpo -c -o xz-list.o `test -f 'list.c' || echo '$(srcdir)/'`list.c
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-list.Tpo $(DEPDIR)/xz-list.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='list.c' object='xz-list.o' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-list.o `test -f 'list.c' || echo '$(srcdir)/'`list.c
+
+xz-list.obj: list.c
+@am__fastdepCC_TRUE@ $(AM_V_CC)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT xz-list.obj -MD -MP -MF $(DEPDIR)/xz-list.Tpo -c -o xz-list.obj `if test -f 'list.c'; then $(CYGPATH_W) 'list.c'; else $(CYGPATH_W) '$(srcdir)/list.c'; fi`
+@am__fastdepCC_TRUE@ $(AM_V_at)$(am__mv) $(DEPDIR)/xz-list.Tpo $(DEPDIR)/xz-list.Po
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ $(AM_V_CC)source='list.c' object='xz-list.obj' libtool=no @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@ $(AM_V_CC@am__nodep@)$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(xz_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o xz-list.obj `if test -f 'list.c'; then $(CYGPATH_W) 'list.c'; else $(CYGPATH_W) '$(srcdir)/list.c'; fi`
+
+mostlyclean-libtool:
+ -rm -f *.lo
+
+clean-libtool:
+ -rm -rf .libs _libs
+install-man1: $(dist_man_MANS)
+ @$(NORMAL_INSTALL)
+ @list1=''; \
+ list2='$(dist_man_MANS)'; \
+ test -n "$(man1dir)" \
+ && test -n "`echo $$list1$$list2`" \
+ || exit 0; \
+ echo " $(MKDIR_P) '$(DESTDIR)$(man1dir)'"; \
+ $(MKDIR_P) "$(DESTDIR)$(man1dir)" || exit 1; \
+ { for i in $$list1; do echo "$$i"; done; \
+ if test -n "$$list2"; then \
+ for i in $$list2; do echo "$$i"; done \
+ | sed -n '/\.1[a-z]*$$/p'; \
+ fi; \
+ } | while read p; do \
+ if test -f $$p; then d=; else d="$(srcdir)/"; fi; \
+ echo "$$d$$p"; echo "$$p"; \
+ done | \
+ sed -e 'n;s,.*/,,;p;h;s,.*\.,,;s,^[^1][0-9a-z]*$$,1,;x' \
+ -e 's,\.[0-9a-z]*$$,,;$(transform);G;s,\n,.,' | \
+ sed 'N;N;s,\n, ,g' | { \
+ list=; while read file base inst; do \
+ if test "$$base" = "$$inst"; then list="$$list $$file"; else \
+ echo " $(INSTALL_DATA) '$$file' '$(DESTDIR)$(man1dir)/$$inst'"; \
+ $(INSTALL_DATA) "$$file" "$(DESTDIR)$(man1dir)/$$inst" || exit $$?; \
+ fi; \
+ done; \
+ for i in $$list; do echo "$$i"; done | $(am__base_list) | \
+ while read files; do \
+ test -z "$$files" || { \
+ echo " $(INSTALL_DATA) $$files '$(DESTDIR)$(man1dir)'"; \
+ $(INSTALL_DATA) $$files "$(DESTDIR)$(man1dir)" || exit $$?; }; \
+ done; }
+
+uninstall-man1:
+ @$(NORMAL_UNINSTALL)
+ @list=''; test -n "$(man1dir)" || exit 0; \
+ files=`{ for i in $$list; do echo "$$i"; done; \
+ l2='$(dist_man_MANS)'; for i in $$l2; do echo "$$i"; done | \
+ sed -n '/\.1[a-z]*$$/p'; \
+ } | sed -e 's,.*/,,;h;s,.*\.,,;s,^[^1][0-9a-z]*$$,1,;x' \
+ -e 's,\.[0-9a-z]*$$,,;$(transform);G;s,\n,.,'`; \
+ dir='$(DESTDIR)$(man1dir)'; $(am__uninstall_files_from_dir)
+
+ID: $(am__tagged_files)
+ $(am__define_uniq_tagged_files); mkid -fID $$unique
+tags: tags-am
+TAGS: tags
+
+tags-am: $(TAGS_DEPENDENCIES) $(am__tagged_files)
+ set x; \
+ here=`pwd`; \
+ $(am__define_uniq_tagged_files); \
+ shift; \
+ if test -z "$(ETAGS_ARGS)$$*$$unique"; then :; else \
+ test -n "$$unique" || unique=$$empty_fix; \
+ if test $$# -gt 0; then \
+ $(ETAGS) $(ETAGSFLAGS) $(AM_ETAGSFLAGS) $(ETAGS_ARGS) \
+ "$$@" $$unique; \
+ else \
+ $(ETAGS) $(ETAGSFLAGS) $(AM_ETAGSFLAGS) $(ETAGS_ARGS) \
+ $$unique; \
+ fi; \
+ fi
+ctags: ctags-am
+
+CTAGS: ctags
+ctags-am: $(TAGS_DEPENDENCIES) $(am__tagged_files)
+ $(am__define_uniq_tagged_files); \
+ test -z "$(CTAGS_ARGS)$$unique" \
+ || $(CTAGS) $(CTAGSFLAGS) $(AM_CTAGSFLAGS) $(CTAGS_ARGS) \
+ $$unique
+
+GTAGS:
+ here=`$(am__cd) $(top_builddir) && pwd` \
+ && $(am__cd) $(top_srcdir) \
+ && gtags -i $(GTAGS_ARGS) "$$here"
+cscopelist: cscopelist-am
+
+cscopelist-am: $(am__tagged_files)
+ list='$(am__tagged_files)'; \
+ case "$(srcdir)" in \
+ [\\/]* | ?:[\\/]*) sdir="$(srcdir)" ;; \
+ *) sdir=$(subdir)/$(srcdir) ;; \
+ esac; \
+ for i in $$list; do \
+ if test -f "$$i"; then \
+ echo "$(subdir)/$$i"; \
+ else \
+ echo "$$sdir/$$i"; \
+ fi; \
+ done >> $(top_builddir)/cscope.files
+
+distclean-tags:
+ -rm -f TAGS ID GTAGS GRTAGS GSYMS GPATH tags
+distdir: $(BUILT_SOURCES)
+ $(MAKE) $(AM_MAKEFLAGS) distdir-am
+
+distdir-am: $(DISTFILES)
+ @srcdirstrip=`echo "$(srcdir)" | sed 's/[].[^$$\\*]/\\\\&/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)/$$file" \
+ || exit 1; \
+ fi; \
+ done
+check-am: all-am
+check: check-am
+all-am: Makefile $(PROGRAMS) $(MANS)
+installdirs:
+ for dir in "$(DESTDIR)$(bindir)" "$(DESTDIR)$(man1dir)"; do \
+ test -z "$$dir" || $(MKDIR_P) "$$dir"; \
+ done
+install: install-am
+install-exec: install-exec-am
+install-data: install-data-am
+uninstall: uninstall-am
+
+install-am: all-am
+ @$(MAKE) $(AM_MAKEFLAGS) install-exec-am install-data-am
+
+installcheck: installcheck-am
+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:
+
+clean-generic:
+
+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)
+
+maintainer-clean-generic:
+ @echo "This command is intended for maintainers to use"
+ @echo "it deletes files that may require special tools to rebuild."
+clean: clean-am
+
+clean-am: clean-binPROGRAMS clean-generic clean-libtool mostlyclean-am
+
+distclean: distclean-am
+ -rm -f ./$(DEPDIR)/xz-args.Po
+ -rm -f ./$(DEPDIR)/xz-coder.Po
+ -rm -f ./$(DEPDIR)/xz-file_io.Po
+ -rm -f ./$(DEPDIR)/xz-hardware.Po
+ -rm -f ./$(DEPDIR)/xz-list.Po
+ -rm -f ./$(DEPDIR)/xz-main.Po
+ -rm -f ./$(DEPDIR)/xz-message.Po
+ -rm -f ./$(DEPDIR)/xz-mytime.Po
+ -rm -f ./$(DEPDIR)/xz-options.Po
+ -rm -f ./$(DEPDIR)/xz-signals.Po
+ -rm -f ./$(DEPDIR)/xz-suffix.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_exit.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_mbstr_fw.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_mbstr_width.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_open_stdxxx.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_progname.Po
+ -rm -f ./$(DEPDIR)/xz-util.Po
+ -rm -f Makefile
+distclean-am: clean-am distclean-compile distclean-generic \
+ distclean-tags
+
+dvi: dvi-am
+
+dvi-am:
+
+html: html-am
+
+html-am:
+
+info: info-am
+
+info-am:
+
+install-data-am: install-man
+ @$(NORMAL_INSTALL)
+ $(MAKE) $(AM_MAKEFLAGS) install-data-hook
+install-dvi: install-dvi-am
+
+install-dvi-am:
+
+install-exec-am: install-binPROGRAMS
+ @$(NORMAL_INSTALL)
+ $(MAKE) $(AM_MAKEFLAGS) install-exec-hook
+install-html: install-html-am
+
+install-html-am:
+
+install-info: install-info-am
+
+install-info-am:
+
+install-man: install-man1
+
+install-pdf: install-pdf-am
+
+install-pdf-am:
+
+install-ps: install-ps-am
+
+install-ps-am:
+
+installcheck-am:
+
+maintainer-clean: maintainer-clean-am
+ -rm -f ./$(DEPDIR)/xz-args.Po
+ -rm -f ./$(DEPDIR)/xz-coder.Po
+ -rm -f ./$(DEPDIR)/xz-file_io.Po
+ -rm -f ./$(DEPDIR)/xz-hardware.Po
+ -rm -f ./$(DEPDIR)/xz-list.Po
+ -rm -f ./$(DEPDIR)/xz-main.Po
+ -rm -f ./$(DEPDIR)/xz-message.Po
+ -rm -f ./$(DEPDIR)/xz-mytime.Po
+ -rm -f ./$(DEPDIR)/xz-options.Po
+ -rm -f ./$(DEPDIR)/xz-signals.Po
+ -rm -f ./$(DEPDIR)/xz-suffix.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_exit.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_mbstr_fw.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_mbstr_width.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_open_stdxxx.Po
+ -rm -f ./$(DEPDIR)/xz-tuklib_progname.Po
+ -rm -f ./$(DEPDIR)/xz-util.Po
+ -rm -f Makefile
+maintainer-clean-am: distclean-am maintainer-clean-generic
+
+mostlyclean: mostlyclean-am
+
+mostlyclean-am: mostlyclean-compile mostlyclean-generic \
+ mostlyclean-libtool
+
+pdf: pdf-am
+
+pdf-am:
+
+ps: ps-am
+
+ps-am:
+
+uninstall-am: uninstall-binPROGRAMS uninstall-man
+ @$(NORMAL_INSTALL)
+ $(MAKE) $(AM_MAKEFLAGS) uninstall-hook
+uninstall-man: uninstall-man1
+
+.MAKE: install-am install-data-am install-exec-am install-strip \
+ uninstall-am
+
+.PHONY: CTAGS GTAGS TAGS all all-am am--depfiles check check-am clean \
+ clean-binPROGRAMS clean-generic clean-libtool cscopelist-am \
+ ctags ctags-am distclean distclean-compile distclean-generic \
+ distclean-libtool distclean-tags distdir dvi dvi-am html \
+ html-am info info-am install install-am install-binPROGRAMS \
+ install-data install-data-am install-data-hook install-dvi \
+ install-dvi-am install-exec install-exec-am install-exec-hook \
+ install-html install-html-am install-info install-info-am \
+ install-man install-man1 install-pdf install-pdf-am install-ps \
+ install-ps-am install-strip installcheck installcheck-am \
+ installdirs maintainer-clean maintainer-clean-generic \
+ mostlyclean mostlyclean-compile mostlyclean-generic \
+ mostlyclean-libtool pdf pdf-am ps ps-am tags tags-am uninstall \
+ uninstall-am uninstall-binPROGRAMS uninstall-hook \
+ uninstall-man uninstall-man1
+
+.PRECIOUS: Makefile
+
+
+# Windows resource compiler support
+.rc.o:
+ $(RC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) \
+ $(xz_CPPFLAGS) $(CPPFLAGS) $(RCFLAGS) -i $< -o $@
+
+install-exec-hook:
+ cd "$(DESTDIR)$(bindir)" && \
+ target=`echo xz | sed '$(transform)'`$(EXEEXT) && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'`$(LN_EXEEXT) && \
+ rm -f "$$link" && \
+ $(LN_S) "$$target" "$$link"; \
+ done
+
+# The installation of translated man pages abuses Automake internals
+# by calling "install-man" with redefined dist_man_MANS and man_MANS.
+# If this breaks some day, don't blame Automake developers.
+install-data-hook:
+ languages= ; \
+ if test "$(USE_NLS)" = yes && test -d "$(top_srcdir)/po4a/man"; then \
+ languages=`ls "$(top_srcdir)/po4a/man"`; \
+ fi; \
+ target=`echo xz | sed '$(transform)'` && \
+ for lang in . $$languages; do \
+ man="$(top_srcdir)/po4a/man/$$lang/xz.1" ; \
+ if test -f "$$man"; then \
+ $(MAKE) dist_man_MANS="$$man" man_MANS= \
+ mandir="$(mandir)/$$lang" install-man; \
+ fi; \
+ man1dir="$(DESTDIR)$(mandir)/$$lang/man1" && \
+ if test -f "$$man1dir/$$target.1"; then ( \
+ cd "$$man1dir" && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f "$$link.1" && \
+ $(LN_S) "$$target.1" "$$link.1"; \
+ done \
+ ); fi; \
+ done
+
+uninstall-hook:
+ cd "$(DESTDIR)$(bindir)" && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'`$(LN_EXEEXT) && \
+ rm -f "$$link"; \
+ done
+ languages= ; \
+ if test "$(USE_NLS)" = yes && test -d "$(top_srcdir)/po4a/man"; then \
+ languages=`ls "$(top_srcdir)/po4a/man"`; \
+ fi; \
+ for lang in . $$languages; do \
+ for name in xz $(xzlinks); do \
+ name=`echo $$name | sed '$(transform)'` && \
+ rm -f "$(DESTDIR)$(mandir)/$$lang/man1/$$name.1"; \
+ done; \
+ 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/src/xz/args.c b/src/xz/args.c
new file mode 100644
index 0000000..17e778c
--- /dev/null
+++ b/src/xz/args.c
@@ -0,0 +1,763 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file args.c
+/// \brief Argument parsing
+///
+/// \note Filter-specific options parsing is in options.c.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+#include "getopt.h"
+#include <ctype.h>
+
+
+bool opt_stdout = false;
+bool opt_force = false;
+bool opt_keep_original = false;
+bool opt_robot = false;
+bool opt_ignore_check = false;
+
+// We don't modify or free() this, but we need to assign it in some
+// non-const pointers.
+const char stdin_filename[] = "(stdin)";
+
+
+/// Parse and set the memory usage limit for compression, decompression,
+/// and/or multithreaded decompression.
+static void
+parse_memlimit(const char *name, const char *name_percentage, const char *str,
+ bool set_compress, bool set_decompress, bool set_mtdec)
+{
+ bool is_percentage = false;
+ uint64_t value;
+
+ const size_t len = strlen(str);
+ if (len > 0 && str[len - 1] == '%') {
+ // Make a copy so that we can get rid of %.
+ //
+ // In the past str wasn't const and we modified it directly
+ // but that modified argv[] and thus affected what was visible
+ // in "ps auxf" or similar tools which was confusing. For
+ // example, --memlimit=50% would show up as --memlimit=50
+ // since the percent sign was overwritten here.
+ char *s = xstrdup(str);
+ s[len - 1] = '\0';
+ is_percentage = true;
+ value = str_to_uint64(name_percentage, s, 1, 100);
+ free(s);
+ } else {
+ // On 32-bit systems, SIZE_MAX would make more sense than
+ // UINT64_MAX. But use UINT64_MAX still so that scripts
+ // that assume > 4 GiB values don't break.
+ value = str_to_uint64(name, str, 0, UINT64_MAX);
+ }
+
+ hardware_memlimit_set(value, set_compress, set_decompress, set_mtdec,
+ is_percentage);
+ return;
+}
+
+
+static void
+parse_block_list(const char *str_const)
+{
+ // We need a modifiable string in the for-loop.
+ char *str_start = xstrdup(str_const);
+ char *str = str_start;
+
+ // It must be non-empty and not begin with a comma.
+ if (str[0] == '\0' || str[0] == ',')
+ message_fatal(_("%s: Invalid argument to --block-list"), str);
+
+ // Count the number of comma-separated strings.
+ size_t count = 1;
+ for (size_t i = 0; str[i] != '\0'; ++i)
+ if (str[i] == ',')
+ ++count;
+
+ // Prevent an unlikely integer overflow.
+ if (count > SIZE_MAX / sizeof(uint64_t) - 1)
+ message_fatal(_("%s: Too many arguments to --block-list"),
+ str);
+
+ // Allocate memory to hold all the sizes specified.
+ // If --block-list was specified already, its value is forgotten.
+ free(opt_block_list);
+ opt_block_list = xmalloc((count + 1) * sizeof(uint64_t));
+
+ for (size_t i = 0; i < count; ++i) {
+ // Locate the next comma and replace it with \0.
+ char *p = strchr(str, ',');
+ if (p != NULL)
+ *p = '\0';
+
+ if (str[0] == '\0') {
+ // There is no string, that is, a comma follows
+ // another comma. Use the previous value.
+ //
+ // NOTE: We checked earlier that the first char
+ // of the whole list cannot be a comma.
+ assert(i > 0);
+ opt_block_list[i] = opt_block_list[i - 1];
+ } else {
+ opt_block_list[i] = str_to_uint64("block-list", str,
+ 0, UINT64_MAX);
+
+ // Zero indicates no more new Blocks.
+ if (opt_block_list[i] == 0) {
+ if (i + 1 != count)
+ message_fatal(_("0 can only be used "
+ "as the last element "
+ "in --block-list"));
+
+ opt_block_list[i] = UINT64_MAX;
+ }
+ }
+
+ str = p + 1;
+ }
+
+ // Terminate the array.
+ opt_block_list[count] = 0;
+
+ free(str_start);
+ return;
+}
+
+
+static void
+parse_real(args_info *args, int argc, char **argv)
+{
+ enum {
+ OPT_X86 = INT_MIN,
+ OPT_POWERPC,
+ OPT_IA64,
+ OPT_ARM,
+ OPT_ARMTHUMB,
+ OPT_ARM64,
+ OPT_SPARC,
+ OPT_DELTA,
+ OPT_LZMA1,
+ OPT_LZMA2,
+
+ OPT_SINGLE_STREAM,
+ OPT_NO_SPARSE,
+ OPT_FILES,
+ OPT_FILES0,
+ OPT_BLOCK_SIZE,
+ OPT_BLOCK_LIST,
+ OPT_MEM_COMPRESS,
+ OPT_MEM_DECOMPRESS,
+ OPT_MEM_MT_DECOMPRESS,
+ OPT_NO_ADJUST,
+ OPT_INFO_MEMORY,
+ OPT_ROBOT,
+ OPT_FLUSH_TIMEOUT,
+ OPT_IGNORE_CHECK,
+ };
+
+ static const char short_opts[]
+ = "cC:defF:hHlkM:qQrS:tT:vVz0123456789";
+
+ static const struct option long_opts[] = {
+ // Operation mode
+ { "compress", no_argument, NULL, 'z' },
+ { "decompress", no_argument, NULL, 'd' },
+ { "uncompress", no_argument, NULL, 'd' },
+ { "test", no_argument, NULL, 't' },
+ { "list", no_argument, NULL, 'l' },
+
+ // Operation modifiers
+ { "keep", no_argument, NULL, 'k' },
+ { "force", no_argument, NULL, 'f' },
+ { "stdout", no_argument, NULL, 'c' },
+ { "to-stdout", no_argument, NULL, 'c' },
+ { "single-stream", no_argument, NULL, OPT_SINGLE_STREAM },
+ { "no-sparse", no_argument, NULL, OPT_NO_SPARSE },
+ { "suffix", required_argument, NULL, 'S' },
+ // { "recursive", no_argument, NULL, 'r' }, // TODO
+ { "files", optional_argument, NULL, OPT_FILES },
+ { "files0", optional_argument, NULL, OPT_FILES0 },
+
+ // Basic compression settings
+ { "format", required_argument, NULL, 'F' },
+ { "check", required_argument, NULL, 'C' },
+ { "ignore-check", no_argument, NULL, OPT_IGNORE_CHECK },
+ { "block-size", required_argument, NULL, OPT_BLOCK_SIZE },
+ { "block-list", required_argument, NULL, OPT_BLOCK_LIST },
+ { "memlimit-compress", required_argument, NULL, OPT_MEM_COMPRESS },
+ { "memlimit-decompress", required_argument, NULL, OPT_MEM_DECOMPRESS },
+ { "memlimit-mt-decompress", required_argument, NULL, OPT_MEM_MT_DECOMPRESS },
+ { "memlimit", required_argument, NULL, 'M' },
+ { "memory", required_argument, NULL, 'M' }, // Old alias
+ { "no-adjust", no_argument, NULL, OPT_NO_ADJUST },
+ { "threads", required_argument, NULL, 'T' },
+ { "flush-timeout", required_argument, NULL, OPT_FLUSH_TIMEOUT },
+
+ { "extreme", no_argument, NULL, 'e' },
+ { "fast", no_argument, NULL, '0' },
+ { "best", no_argument, NULL, '9' },
+
+ // Filters
+ { "lzma1", optional_argument, NULL, OPT_LZMA1 },
+ { "lzma2", optional_argument, NULL, OPT_LZMA2 },
+ { "x86", optional_argument, NULL, OPT_X86 },
+ { "powerpc", optional_argument, NULL, OPT_POWERPC },
+ { "ia64", optional_argument, NULL, OPT_IA64 },
+ { "arm", optional_argument, NULL, OPT_ARM },
+ { "armthumb", optional_argument, NULL, OPT_ARMTHUMB },
+ { "arm64", optional_argument, NULL, OPT_ARM64 },
+ { "sparc", optional_argument, NULL, OPT_SPARC },
+ { "delta", optional_argument, NULL, OPT_DELTA },
+
+ // Other options
+ { "quiet", no_argument, NULL, 'q' },
+ { "verbose", no_argument, NULL, 'v' },
+ { "no-warn", no_argument, NULL, 'Q' },
+ { "robot", no_argument, NULL, OPT_ROBOT },
+ { "info-memory", no_argument, NULL, OPT_INFO_MEMORY },
+ { "help", no_argument, NULL, 'h' },
+ { "long-help", no_argument, NULL, 'H' },
+ { "version", no_argument, NULL, 'V' },
+
+ { NULL, 0, NULL, 0 }
+ };
+
+ int c;
+
+ while ((c = getopt_long(argc, argv, short_opts, long_opts, NULL))
+ != -1) {
+ switch (c) {
+ // Compression preset (also for decompression if --format=raw)
+ case '0': case '1': case '2': case '3': case '4':
+ case '5': case '6': case '7': case '8': case '9':
+ coder_set_preset((uint32_t)(c - '0'));
+ break;
+
+ // --memlimit-compress
+ case OPT_MEM_COMPRESS:
+ parse_memlimit("memlimit-compress",
+ "memlimit-compress%", optarg,
+ true, false, false);
+ break;
+
+ // --memlimit-decompress
+ case OPT_MEM_DECOMPRESS:
+ parse_memlimit("memlimit-decompress",
+ "memlimit-decompress%", optarg,
+ false, true, false);
+ break;
+
+ // --memlimit-mt-decompress
+ case OPT_MEM_MT_DECOMPRESS:
+ parse_memlimit("memlimit-mt-decompress",
+ "memlimit-mt-decompress%", optarg,
+ false, false, true);
+ break;
+
+ // --memlimit
+ case 'M':
+ parse_memlimit("memlimit", "memlimit%", optarg,
+ true, true, true);
+ break;
+
+ // --suffix
+ case 'S':
+ suffix_set(optarg);
+ break;
+
+ case 'T': {
+ // Since xz 5.4.0: Ignore leading '+' first.
+ const char *s = optarg;
+ if (optarg[0] == '+')
+ ++s;
+
+ // The max is from src/liblzma/common/common.h.
+ uint32_t t = str_to_uint64("threads", s, 0, 16384);
+
+ // If leading '+' was used then use multi-threaded
+ // mode even if exactly one thread was specified.
+ if (t == 1 && optarg[0] == '+')
+ t = UINT32_MAX;
+
+ hardware_threads_set(t);
+ break;
+ }
+
+ // --version
+ case 'V':
+ // This doesn't return.
+ message_version();
+
+ // --stdout
+ case 'c':
+ opt_stdout = true;
+ break;
+
+ // --decompress
+ case 'd':
+ opt_mode = MODE_DECOMPRESS;
+ break;
+
+ // --extreme
+ case 'e':
+ coder_set_extreme();
+ break;
+
+ // --force
+ case 'f':
+ opt_force = true;
+ break;
+
+ // --info-memory
+ case OPT_INFO_MEMORY:
+ // This doesn't return.
+ hardware_memlimit_show();
+
+ // --help
+ case 'h':
+ // This doesn't return.
+ message_help(false);
+
+ // --long-help
+ case 'H':
+ // This doesn't return.
+ message_help(true);
+
+ // --list
+ case 'l':
+ opt_mode = MODE_LIST;
+ break;
+
+ // --keep
+ case 'k':
+ opt_keep_original = true;
+ break;
+
+ // --quiet
+ case 'q':
+ message_verbosity_decrease();
+ break;
+
+ case 'Q':
+ set_exit_no_warn();
+ break;
+
+ case 't':
+ opt_mode = MODE_TEST;
+ break;
+
+ // --verbose
+ case 'v':
+ message_verbosity_increase();
+ break;
+
+ // --robot
+ case OPT_ROBOT:
+ opt_robot = true;
+
+ // This is to make sure that floating point numbers
+ // always have a dot as decimal separator.
+ setlocale(LC_NUMERIC, "C");
+ break;
+
+ case 'z':
+ opt_mode = MODE_COMPRESS;
+ break;
+
+ // Filter setup
+
+ case OPT_X86:
+ coder_add_filter(LZMA_FILTER_X86,
+ options_bcj(optarg));
+ break;
+
+ case OPT_POWERPC:
+ coder_add_filter(LZMA_FILTER_POWERPC,
+ options_bcj(optarg));
+ break;
+
+ case OPT_IA64:
+ coder_add_filter(LZMA_FILTER_IA64,
+ options_bcj(optarg));
+ break;
+
+ case OPT_ARM:
+ coder_add_filter(LZMA_FILTER_ARM,
+ options_bcj(optarg));
+ break;
+
+ case OPT_ARMTHUMB:
+ coder_add_filter(LZMA_FILTER_ARMTHUMB,
+ options_bcj(optarg));
+ break;
+
+ case OPT_ARM64:
+ coder_add_filter(LZMA_FILTER_ARM64,
+ options_bcj(optarg));
+ break;
+
+ case OPT_SPARC:
+ coder_add_filter(LZMA_FILTER_SPARC,
+ options_bcj(optarg));
+ break;
+
+ case OPT_DELTA:
+ coder_add_filter(LZMA_FILTER_DELTA,
+ options_delta(optarg));
+ break;
+
+ case OPT_LZMA1:
+ coder_add_filter(LZMA_FILTER_LZMA1,
+ options_lzma(optarg));
+ break;
+
+ case OPT_LZMA2:
+ coder_add_filter(LZMA_FILTER_LZMA2,
+ options_lzma(optarg));
+ break;
+
+ // Other
+
+ // --format
+ case 'F': {
+ // Just in case, support both "lzma" and "alone" since
+ // the latter was used for forward compatibility in
+ // LZMA Utils 4.32.x.
+ static const struct {
+ char str[8];
+ enum format_type format;
+ } types[] = {
+ { "auto", FORMAT_AUTO },
+ { "xz", FORMAT_XZ },
+ { "lzma", FORMAT_LZMA },
+ { "alone", FORMAT_LZMA },
+#ifdef HAVE_LZIP_DECODER
+ { "lzip", FORMAT_LZIP },
+#endif
+ { "raw", FORMAT_RAW },
+ };
+
+ size_t i = 0;
+ while (strcmp(types[i].str, optarg) != 0)
+ if (++i == ARRAY_SIZE(types))
+ message_fatal(_("%s: Unknown file "
+ "format type"),
+ optarg);
+
+ opt_format = types[i].format;
+ break;
+ }
+
+ // --check
+ case 'C': {
+ static const struct {
+ char str[8];
+ lzma_check check;
+ } types[] = {
+ { "none", LZMA_CHECK_NONE },
+ { "crc32", LZMA_CHECK_CRC32 },
+ { "crc64", LZMA_CHECK_CRC64 },
+ { "sha256", LZMA_CHECK_SHA256 },
+ };
+
+ size_t i = 0;
+ while (strcmp(types[i].str, optarg) != 0) {
+ if (++i == ARRAY_SIZE(types))
+ message_fatal(_("%s: Unsupported "
+ "integrity "
+ "check type"), optarg);
+ }
+
+ // Use a separate check in case we are using different
+ // liblzma than what was used to compile us.
+ if (!lzma_check_is_supported(types[i].check))
+ message_fatal(_("%s: Unsupported integrity "
+ "check type"), optarg);
+
+ coder_set_check(types[i].check);
+ break;
+ }
+
+ case OPT_IGNORE_CHECK:
+ opt_ignore_check = true;
+ break;
+
+ case OPT_BLOCK_SIZE:
+ opt_block_size = str_to_uint64("block-size", optarg,
+ 0, LZMA_VLI_MAX);
+ break;
+
+ case OPT_BLOCK_LIST: {
+ parse_block_list(optarg);
+ break;
+ }
+
+ case OPT_SINGLE_STREAM:
+ opt_single_stream = true;
+ break;
+
+ case OPT_NO_SPARSE:
+ io_no_sparse();
+ break;
+
+ case OPT_FILES:
+ args->files_delim = '\n';
+
+ // Fall through
+
+ case OPT_FILES0:
+ if (args->files_name != NULL)
+ message_fatal(_("Only one file can be "
+ "specified with `--files' "
+ "or `--files0'."));
+
+ if (optarg == NULL) {
+ args->files_name = stdin_filename;
+ args->files_file = stdin;
+ } else {
+ args->files_name = optarg;
+ args->files_file = fopen(optarg,
+ c == OPT_FILES ? "r" : "rb");
+ if (args->files_file == NULL)
+ // TRANSLATORS: This is a translatable
+ // string because French needs a space
+ // before the colon ("%s : %s").
+ message_fatal(_("%s: %s"), optarg,
+ strerror(errno));
+ }
+
+ break;
+
+ case OPT_NO_ADJUST:
+ opt_auto_adjust = false;
+ break;
+
+ case OPT_FLUSH_TIMEOUT:
+ opt_flush_timeout = str_to_uint64("flush-timeout",
+ optarg, 0, UINT64_MAX);
+ break;
+
+ default:
+ message_try_help();
+ tuklib_exit(E_ERROR, E_ERROR, false);
+ }
+ }
+
+ return;
+}
+
+
+static void
+parse_environment(args_info *args, char *argv0, const char *varname)
+{
+ char *env = getenv(varname);
+ if (env == NULL)
+ return;
+
+ // We modify the string, so make a copy of it.
+ env = xstrdup(env);
+
+ // Calculate the number of arguments in env. argc stats at one
+ // to include space for the program name.
+ int argc = 1;
+ bool prev_was_space = true;
+ for (size_t i = 0; env[i] != '\0'; ++i) {
+ // NOTE: Cast to unsigned char is needed so that correct
+ // value gets passed to isspace(), which expects
+ // unsigned char cast to int. Casting to int is done
+ // automatically due to integer promotion, but we need to
+ // force char to unsigned char manually. Otherwise 8-bit
+ // characters would get promoted to wrong value if
+ // char is signed.
+ if (isspace((unsigned char)env[i])) {
+ prev_was_space = true;
+ } else if (prev_was_space) {
+ prev_was_space = false;
+
+ // Keep argc small enough to fit into a signed int
+ // and to keep it usable for memory allocation.
+ if (++argc == my_min(
+ INT_MAX, SIZE_MAX / sizeof(char *)))
+ message_fatal(_("The environment variable "
+ "%s contains too many "
+ "arguments"), varname);
+ }
+ }
+
+ // Allocate memory to hold pointers to the arguments. Add one to get
+ // space for the terminating NULL (if some systems happen to need it).
+ char **argv = xmalloc(((size_t)(argc) + 1) * sizeof(char *));
+ argv[0] = argv0;
+ argv[argc] = NULL;
+
+ // Go through the string again. Split the arguments using '\0'
+ // characters and add pointers to the resulting strings to argv.
+ argc = 1;
+ prev_was_space = true;
+ for (size_t i = 0; env[i] != '\0'; ++i) {
+ if (isspace((unsigned char)env[i])) {
+ prev_was_space = true;
+ env[i] = '\0';
+ } else if (prev_was_space) {
+ prev_was_space = false;
+ argv[argc++] = env + i;
+ }
+ }
+
+ // Parse the argument list we got from the environment. All non-option
+ // arguments i.e. filenames are ignored.
+ parse_real(args, argc, argv);
+
+ // Reset the state of the getopt_long() so that we can parse the
+ // command line options too. There are two incompatible ways to
+ // do it.
+#ifdef HAVE_OPTRESET
+ // BSD
+ optind = 1;
+ optreset = 1;
+#else
+ // GNU, Solaris
+ optind = 0;
+#endif
+
+ // We don't need the argument list from environment anymore.
+ free(argv);
+ free(env);
+
+ return;
+}
+
+
+extern void
+args_parse(args_info *args, int argc, char **argv)
+{
+ // Initialize those parts of *args that we need later.
+ args->files_name = NULL;
+ args->files_file = NULL;
+ args->files_delim = '\0';
+
+ // Check how we were called.
+ {
+ // Remove the leading path name, if any.
+ const char *name = strrchr(argv[0], '/');
+ if (name == NULL)
+ name = argv[0];
+ else
+ ++name;
+
+ // NOTE: It's possible that name[0] is now '\0' if argv[0]
+ // is weird, but it doesn't matter here.
+
+ // Look for full command names instead of substrings like
+ // "un", "cat", and "lz" to reduce possibility of false
+ // positives when the programs have been renamed.
+ if (strstr(name, "xzcat") != NULL) {
+ opt_mode = MODE_DECOMPRESS;
+ opt_stdout = true;
+ } else if (strstr(name, "unxz") != NULL) {
+ opt_mode = MODE_DECOMPRESS;
+ } else if (strstr(name, "lzcat") != NULL) {
+ opt_format = FORMAT_LZMA;
+ opt_mode = MODE_DECOMPRESS;
+ opt_stdout = true;
+ } else if (strstr(name, "unlzma") != NULL) {
+ opt_format = FORMAT_LZMA;
+ opt_mode = MODE_DECOMPRESS;
+ } else if (strstr(name, "lzma") != NULL) {
+ opt_format = FORMAT_LZMA;
+ }
+ }
+
+ // First the flags from the environment
+ parse_environment(args, argv[0], "XZ_DEFAULTS");
+ parse_environment(args, argv[0], "XZ_OPT");
+
+ // Then from the command line
+ parse_real(args, argc, argv);
+
+ // If encoder or decoder support was omitted at build time,
+ // show an error now so that the rest of the code can rely on
+ // that whatever is in opt_mode is also supported.
+#ifndef HAVE_ENCODERS
+ if (opt_mode == MODE_COMPRESS)
+ message_fatal(_("Compression support was disabled "
+ "at build time"));
+#endif
+#ifndef HAVE_DECODERS
+ // Even MODE_LIST cannot work without decoder support so MODE_COMPRESS
+ // is the only valid choice.
+ if (opt_mode != MODE_COMPRESS)
+ message_fatal(_("Decompression support was disabled "
+ "at build time"));
+#endif
+
+#ifdef HAVE_LZIP_DECODER
+ if (opt_mode == MODE_COMPRESS && opt_format == FORMAT_LZIP)
+ message_fatal(_("Compression of lzip files (.lz) "
+ "is not supported"));
+#endif
+
+ // Never remove the source file when the destination is not on disk.
+ // In test mode the data is written nowhere, but setting opt_stdout
+ // will make the rest of the code behave well.
+ if (opt_stdout || opt_mode == MODE_TEST) {
+ opt_keep_original = true;
+ opt_stdout = true;
+ }
+
+ // When compressing, if no --format flag was used, or it
+ // was --format=auto, we compress to the .xz format.
+ if (opt_mode == MODE_COMPRESS && opt_format == FORMAT_AUTO)
+ opt_format = FORMAT_XZ;
+
+ // Compression settings need to be validated (options themselves and
+ // their memory usage) when compressing to any file format. It has to
+ // be done also when uncompressing raw data, since for raw decoding
+ // the options given on the command line are used to know what kind
+ // of raw data we are supposed to decode.
+ if (opt_mode == MODE_COMPRESS || (opt_format == FORMAT_RAW
+ && opt_mode != MODE_LIST))
+ coder_set_compression_settings();
+
+ // If raw format is used and a custom suffix is not provided,
+ // then only stdout mode can be used when compressing or decompressing.
+ if (opt_format == FORMAT_RAW && !suffix_is_set() && !opt_stdout
+ && (opt_mode == MODE_COMPRESS
+ || opt_mode == MODE_DECOMPRESS))
+ message_fatal(_("With --format=raw, --suffix=.SUF is "
+ "required unless writing to stdout"));
+
+ // If no filenames are given, use stdin.
+ if (argv[optind] == NULL && args->files_name == NULL) {
+ // We don't modify or free() the "-" constant. The caller
+ // modifies this so don't make the struct itself const.
+ static char *names_stdin[2] = { (char *)"-", NULL };
+ args->arg_names = names_stdin;
+ args->arg_count = 1;
+ } else {
+ // We got at least one filename from the command line, or
+ // --files or --files0 was specified.
+ args->arg_names = argv + optind;
+ args->arg_count = (unsigned int)(argc - optind);
+ }
+
+ return;
+}
+
+
+#ifndef NDEBUG
+extern void
+args_free(void)
+{
+ free(opt_block_list);
+ return;
+}
+#endif
diff --git a/src/xz/args.h b/src/xz/args.h
new file mode 100644
index 0000000..a1a5930
--- /dev/null
+++ b/src/xz/args.h
@@ -0,0 +1,44 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file args.h
+/// \brief Argument parsing
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+typedef struct {
+ /// Filenames from command line
+ char **arg_names;
+
+ /// Number of filenames from command line
+ unsigned int arg_count;
+
+ /// Name of the file from which to read filenames. This is NULL
+ /// if --files or --files0 was not used.
+ const char *files_name;
+
+ /// File opened for reading from which filenames are read. This is
+ /// non-NULL only if files_name is non-NULL.
+ FILE *files_file;
+
+ /// Delimiter for filenames read from files_file
+ char files_delim;
+
+} args_info;
+
+
+extern bool opt_stdout;
+extern bool opt_force;
+extern bool opt_keep_original;
+// extern bool opt_recursive;
+extern bool opt_robot;
+extern bool opt_ignore_check;
+
+extern const char stdin_filename[];
+
+extern void args_parse(args_info *args, int argc, char **argv);
+extern void args_free(void);
diff --git a/src/xz/coder.c b/src/xz/coder.c
new file mode 100644
index 0000000..91d40ed
--- /dev/null
+++ b/src/xz/coder.c
@@ -0,0 +1,1109 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file coder.c
+/// \brief Compresses or uncompresses a file
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+
+/// Return value type for coder_init().
+enum coder_init_ret {
+ CODER_INIT_NORMAL,
+ CODER_INIT_PASSTHRU,
+ CODER_INIT_ERROR,
+};
+
+
+enum operation_mode opt_mode = MODE_COMPRESS;
+enum format_type opt_format = FORMAT_AUTO;
+bool opt_auto_adjust = true;
+bool opt_single_stream = false;
+uint64_t opt_block_size = 0;
+uint64_t *opt_block_list = NULL;
+
+
+/// Stream used to communicate with liblzma
+static lzma_stream strm = LZMA_STREAM_INIT;
+
+/// Filters needed for all encoding all formats, and also decoding in raw data
+static lzma_filter filters[LZMA_FILTERS_MAX + 1];
+
+/// Input and output buffers
+static io_buf in_buf;
+static io_buf out_buf;
+
+/// Number of filters. Zero indicates that we are using a preset.
+static uint32_t filters_count = 0;
+
+/// Number of the preset (0-9)
+static uint32_t preset_number = LZMA_PRESET_DEFAULT;
+
+/// Integrity check type
+static lzma_check check;
+
+/// This becomes false if the --check=CHECK option is used.
+static bool check_default = true;
+
+/// Indicates if unconsumed input is allowed to remain after
+/// decoding has successfully finished. This is set for each file
+/// in coder_init().
+static bool allow_trailing_input;
+
+#ifdef MYTHREAD_ENABLED
+static lzma_mt mt_options = {
+ .flags = 0,
+ .timeout = 300,
+ .filters = filters,
+};
+#endif
+
+
+extern void
+coder_set_check(lzma_check new_check)
+{
+ check = new_check;
+ check_default = false;
+ return;
+}
+
+
+static void
+forget_filter_chain(void)
+{
+ // Setting a preset makes us forget a possibly defined custom
+ // filter chain.
+ while (filters_count > 0) {
+ --filters_count;
+ free(filters[filters_count].options);
+ filters[filters_count].options = NULL;
+ }
+
+ return;
+}
+
+
+extern void
+coder_set_preset(uint32_t new_preset)
+{
+ preset_number &= ~LZMA_PRESET_LEVEL_MASK;
+ preset_number |= new_preset;
+ forget_filter_chain();
+ return;
+}
+
+
+extern void
+coder_set_extreme(void)
+{
+ preset_number |= LZMA_PRESET_EXTREME;
+ forget_filter_chain();
+ return;
+}
+
+
+extern void
+coder_add_filter(lzma_vli id, void *options)
+{
+ if (filters_count == LZMA_FILTERS_MAX)
+ message_fatal(_("Maximum number of filters is four"));
+
+ filters[filters_count].id = id;
+ filters[filters_count].options = options;
+ ++filters_count;
+
+ // Setting a custom filter chain makes us forget the preset options.
+ // This makes a difference if one specifies e.g. "xz -9 --lzma2 -e"
+ // where the custom filter chain resets the preset level back to
+ // the default 6, making the example equivalent to "xz -6e".
+ preset_number = LZMA_PRESET_DEFAULT;
+
+ return;
+}
+
+
+tuklib_attr_noreturn
+static void
+memlimit_too_small(uint64_t memory_usage)
+{
+ message(V_ERROR, _("Memory usage limit is too low for the given "
+ "filter setup."));
+ message_mem_needed(V_ERROR, memory_usage);
+ tuklib_exit(E_ERROR, E_ERROR, false);
+}
+
+
+extern void
+coder_set_compression_settings(void)
+{
+#ifdef HAVE_LZIP_DECODER
+ // .lz compression isn't supported.
+ assert(opt_format != FORMAT_LZIP);
+#endif
+
+ // The default check type is CRC64, but fallback to CRC32
+ // if CRC64 isn't supported by the copy of liblzma we are
+ // using. CRC32 is always supported.
+ if (check_default) {
+ check = LZMA_CHECK_CRC64;
+ if (!lzma_check_is_supported(check))
+ check = LZMA_CHECK_CRC32;
+ }
+
+ // Options for LZMA1 or LZMA2 in case we are using a preset.
+ static lzma_options_lzma opt_lzma;
+
+ if (filters_count == 0) {
+ // We are using a preset. This is not a good idea in raw mode
+ // except when playing around with things. Different versions
+ // of this software may use different options in presets, and
+ // thus make uncompressing the raw data difficult.
+ if (opt_format == FORMAT_RAW) {
+ // The message is shown only if warnings are allowed
+ // but the exit status isn't changed.
+ message(V_WARNING, _("Using a preset in raw mode "
+ "is discouraged."));
+ message(V_WARNING, _("The exact options of the "
+ "presets may vary between software "
+ "versions."));
+ }
+
+ // Get the preset for LZMA1 or LZMA2.
+ if (lzma_lzma_preset(&opt_lzma, preset_number))
+ message_bug();
+
+ // Use LZMA2 except with --format=lzma we use LZMA1.
+ filters[0].id = opt_format == FORMAT_LZMA
+ ? LZMA_FILTER_LZMA1 : LZMA_FILTER_LZMA2;
+ filters[0].options = &opt_lzma;
+ filters_count = 1;
+ }
+
+ // Terminate the filter options array.
+ filters[filters_count].id = LZMA_VLI_UNKNOWN;
+
+ // If we are using the .lzma format, allow exactly one filter
+ // which has to be LZMA1.
+ if (opt_format == FORMAT_LZMA && (filters_count != 1
+ || filters[0].id != LZMA_FILTER_LZMA1))
+ message_fatal(_("The .lzma format supports only "
+ "the LZMA1 filter"));
+
+ // If we are using the .xz format, make sure that there is no LZMA1
+ // filter to prevent LZMA_PROG_ERROR.
+ if (opt_format == FORMAT_XZ)
+ for (size_t i = 0; i < filters_count; ++i)
+ if (filters[i].id == LZMA_FILTER_LZMA1)
+ message_fatal(_("LZMA1 cannot be used "
+ "with the .xz format"));
+
+ // Print the selected filter chain.
+ message_filters_show(V_DEBUG, filters);
+
+ // The --flush-timeout option requires LZMA_SYNC_FLUSH support
+ // from the filter chain. Currently threaded encoder doesn't support
+ // LZMA_SYNC_FLUSH so single-threaded mode must be used.
+ if (opt_mode == MODE_COMPRESS && opt_flush_timeout != 0) {
+ for (size_t i = 0; i < filters_count; ++i) {
+ switch (filters[i].id) {
+ case LZMA_FILTER_LZMA2:
+ case LZMA_FILTER_DELTA:
+ break;
+
+ default:
+ message_fatal(_("The filter chain is "
+ "incompatible with --flush-timeout"));
+ }
+ }
+
+ if (hardware_threads_is_mt()) {
+ message(V_WARNING, _("Switching to single-threaded "
+ "mode due to --flush-timeout"));
+ hardware_threads_set(1);
+ }
+ }
+
+ // Get the memory usage. Note that if --format=raw was used,
+ // we can be decompressing.
+ //
+ // If multithreaded .xz compression is done, this value will be
+ // replaced.
+ uint64_t memory_limit = hardware_memlimit_get(opt_mode);
+ uint64_t memory_usage = UINT64_MAX;
+ if (opt_mode == MODE_COMPRESS) {
+#ifdef HAVE_ENCODERS
+# ifdef MYTHREAD_ENABLED
+ if (opt_format == FORMAT_XZ && hardware_threads_is_mt()) {
+ memory_limit = hardware_memlimit_mtenc_get();
+ mt_options.threads = hardware_threads_get();
+ mt_options.block_size = opt_block_size;
+ mt_options.check = check;
+ memory_usage = lzma_stream_encoder_mt_memusage(
+ &mt_options);
+ if (memory_usage != UINT64_MAX)
+ message(V_DEBUG, _("Using up to %" PRIu32
+ " threads."),
+ mt_options.threads);
+ } else
+# endif
+ {
+ memory_usage = lzma_raw_encoder_memusage(filters);
+ }
+#endif
+ } else {
+#ifdef HAVE_DECODERS
+ memory_usage = lzma_raw_decoder_memusage(filters);
+#endif
+ }
+
+ if (memory_usage == UINT64_MAX)
+ message_fatal(_("Unsupported filter chain or filter options"));
+
+ // Print memory usage info before possible dictionary
+ // size auto-adjusting.
+ //
+ // NOTE: If only encoder support was built, we cannot show the
+ // what the decoder memory usage will be.
+ message_mem_needed(V_DEBUG, memory_usage);
+#ifdef HAVE_DECODERS
+ if (opt_mode == MODE_COMPRESS) {
+ const uint64_t decmem = lzma_raw_decoder_memusage(filters);
+ if (decmem != UINT64_MAX)
+ message(V_DEBUG, _("Decompression will need "
+ "%s MiB of memory."), uint64_to_str(
+ round_up_to_mib(decmem), 0));
+ }
+#endif
+
+ if (memory_usage <= memory_limit)
+ return;
+
+ // With --format=raw settings are never adjusted to meet
+ // the memory usage limit.
+ if (opt_format == FORMAT_RAW)
+ memlimit_too_small(memory_usage);
+
+ assert(opt_mode == MODE_COMPRESS);
+
+#ifdef HAVE_ENCODERS
+# ifdef MYTHREAD_ENABLED
+ if (opt_format == FORMAT_XZ && hardware_threads_is_mt()) {
+ // Try to reduce the number of threads before
+ // adjusting the compression settings down.
+ while (mt_options.threads > 1) {
+ // Reduce the number of threads by one and check
+ // the memory usage.
+ --mt_options.threads;
+ memory_usage = lzma_stream_encoder_mt_memusage(
+ &mt_options);
+ if (memory_usage == UINT64_MAX)
+ message_bug();
+
+ if (memory_usage <= memory_limit) {
+ // The memory usage is now low enough.
+ message(V_WARNING, _("Reduced the number of "
+ "threads from %s to %s to not exceed "
+ "the memory usage limit of %s MiB"),
+ uint64_to_str(
+ hardware_threads_get(), 0),
+ uint64_to_str(mt_options.threads, 1),
+ uint64_to_str(round_up_to_mib(
+ memory_limit), 2));
+ return;
+ }
+ }
+
+ // If the memory usage limit is only a soft limit (automatic
+ // number of threads and no --memlimit-compress), the limit
+ // is only used to reduce the number of threads and once at
+ // just one thread, the limit is completely ignored. This
+ // way -T0 won't use insane amount of memory but at the same
+ // time the soft limit will never make xz fail and never make
+ // xz change settings that would affect the compressed output.
+ if (hardware_memlimit_mtenc_is_default()) {
+ message(V_WARNING, _("Reduced the number of threads "
+ "from %s to one. The automatic memory usage "
+ "limit of %s MiB is still being exceeded. "
+ "%s MiB of memory is required. "
+ "Continuing anyway."),
+ uint64_to_str(hardware_threads_get(), 0),
+ uint64_to_str(
+ round_up_to_mib(memory_limit), 1),
+ uint64_to_str(
+ round_up_to_mib(memory_usage), 2));
+ return;
+ }
+
+ // If --no-adjust was used, we cannot drop to single-threaded
+ // mode since it produces different compressed output.
+ //
+ // NOTE: In xz 5.2.x, --no-adjust also prevented reducing
+ // the number of threads. This changed in 5.3.3alpha.
+ if (!opt_auto_adjust)
+ memlimit_too_small(memory_usage);
+
+ // Switch to single-threaded mode. It uses
+ // less memory than using one thread in
+ // the multithreaded mode but the output
+ // is also different.
+ hardware_threads_set(1);
+ memory_usage = lzma_raw_encoder_memusage(filters);
+ message(V_WARNING, _("Switching to single-threaded mode "
+ "to not exceed the memory usage limit of %s MiB"),
+ uint64_to_str(round_up_to_mib(memory_limit), 0));
+ }
+# endif
+
+ if (memory_usage <= memory_limit)
+ return;
+
+ // Don't adjust LZMA2 or LZMA1 dictionary size if --no-adjust
+ // was specified as that would change the compressed output.
+ if (!opt_auto_adjust)
+ memlimit_too_small(memory_usage);
+
+ // Look for the last filter if it is LZMA2 or LZMA1, so we can make
+ // it use less RAM. With other filters we don't know what to do.
+ size_t i = 0;
+ while (filters[i].id != LZMA_FILTER_LZMA2
+ && filters[i].id != LZMA_FILTER_LZMA1) {
+ if (filters[i].id == LZMA_VLI_UNKNOWN)
+ memlimit_too_small(memory_usage);
+
+ ++i;
+ }
+
+ // Decrease the dictionary size until we meet the memory
+ // usage limit. First round down to full mebibytes.
+ lzma_options_lzma *opt = filters[i].options;
+ const uint32_t orig_dict_size = opt->dict_size;
+ opt->dict_size &= ~((UINT32_C(1) << 20) - 1);
+ while (true) {
+ // If it is below 1 MiB, auto-adjusting failed. We could be
+ // more sophisticated and scale it down even more, but let's
+ // see if many complain about this version.
+ //
+ // FIXME: Displays the scaled memory usage instead
+ // of the original.
+ if (opt->dict_size < (UINT32_C(1) << 20))
+ memlimit_too_small(memory_usage);
+
+ memory_usage = lzma_raw_encoder_memusage(filters);
+ if (memory_usage == UINT64_MAX)
+ message_bug();
+
+ // Accept it if it is low enough.
+ if (memory_usage <= memory_limit)
+ break;
+
+ // Otherwise 1 MiB down and try again. I hope this
+ // isn't too slow method for cases where the original
+ // dict_size is very big.
+ opt->dict_size -= UINT32_C(1) << 20;
+ }
+
+ // Tell the user that we decreased the dictionary size.
+ message(V_WARNING, _("Adjusted LZMA%c dictionary size "
+ "from %s MiB to %s MiB to not exceed "
+ "the memory usage limit of %s MiB"),
+ filters[i].id == LZMA_FILTER_LZMA2
+ ? '2' : '1',
+ uint64_to_str(orig_dict_size >> 20, 0),
+ uint64_to_str(opt->dict_size >> 20, 1),
+ uint64_to_str(round_up_to_mib(memory_limit), 2));
+#endif
+
+ return;
+}
+
+
+#ifdef HAVE_DECODERS
+/// Return true if the data in in_buf seems to be in the .xz format.
+static bool
+is_format_xz(void)
+{
+ // Specify the magic as hex to be compatible with EBCDIC systems.
+ static const uint8_t magic[6] = { 0xFD, 0x37, 0x7A, 0x58, 0x5A, 0x00 };
+ return strm.avail_in >= sizeof(magic)
+ && memcmp(in_buf.u8, magic, sizeof(magic)) == 0;
+}
+
+
+/// Return true if the data in in_buf seems to be in the .lzma format.
+static bool
+is_format_lzma(void)
+{
+ // The .lzma header is 13 bytes.
+ if (strm.avail_in < 13)
+ return false;
+
+ // Decode the LZMA1 properties.
+ lzma_filter filter = { .id = LZMA_FILTER_LZMA1 };
+ if (lzma_properties_decode(&filter, NULL, in_buf.u8, 5) != LZMA_OK)
+ return false;
+
+ // A hack to ditch tons of false positives: We allow only dictionary
+ // sizes that are 2^n or 2^n + 2^(n-1) or UINT32_MAX. LZMA_Alone
+ // created only files with 2^n, but accepts any dictionary size.
+ // If someone complains, this will be reconsidered.
+ lzma_options_lzma *opt = filter.options;
+ const uint32_t dict_size = opt->dict_size;
+ free(opt);
+
+ if (dict_size != UINT32_MAX) {
+ uint32_t d = dict_size - 1;
+ d |= d >> 2;
+ d |= d >> 3;
+ d |= d >> 4;
+ d |= d >> 8;
+ d |= d >> 16;
+ ++d;
+ if (d != dict_size || dict_size == 0)
+ return false;
+ }
+
+ // Another hack to ditch false positives: Assume that if the
+ // uncompressed size is known, it must be less than 256 GiB.
+ // Again, if someone complains, this will be reconsidered.
+ uint64_t uncompressed_size = 0;
+ for (size_t i = 0; i < 8; ++i)
+ uncompressed_size |= (uint64_t)(in_buf.u8[5 + i]) << (i * 8);
+
+ if (uncompressed_size != UINT64_MAX
+ && uncompressed_size > (UINT64_C(1) << 38))
+ return false;
+
+ return true;
+}
+
+
+#ifdef HAVE_LZIP_DECODER
+/// Return true if the data in in_buf seems to be in the .lz format.
+static bool
+is_format_lzip(void)
+{
+ static const uint8_t magic[4] = { 0x4C, 0x5A, 0x49, 0x50 };
+ return strm.avail_in >= sizeof(magic)
+ && memcmp(in_buf.u8, magic, sizeof(magic)) == 0;
+}
+#endif
+#endif
+
+
+/// Detect the input file type (for now, this done only when decompressing),
+/// and initialize an appropriate coder. Return value indicates if a normal
+/// liblzma-based coder was initialized (CODER_INIT_NORMAL), if passthru
+/// mode should be used (CODER_INIT_PASSTHRU), or if an error occurred
+/// (CODER_INIT_ERROR).
+static enum coder_init_ret
+coder_init(file_pair *pair)
+{
+ lzma_ret ret = LZMA_PROG_ERROR;
+
+ // In most cases if there is input left when coding finishes,
+ // something has gone wrong. Exceptions are --single-stream
+ // and decoding .lz files which can contain trailing non-.lz data.
+ // These will be handled later in this function.
+ allow_trailing_input = false;
+
+ if (opt_mode == MODE_COMPRESS) {
+#ifdef HAVE_ENCODERS
+ switch (opt_format) {
+ case FORMAT_AUTO:
+ // args.c ensures this.
+ assert(0);
+ break;
+
+ case FORMAT_XZ:
+# ifdef MYTHREAD_ENABLED
+ if (hardware_threads_is_mt())
+ ret = lzma_stream_encoder_mt(
+ &strm, &mt_options);
+ else
+# endif
+ ret = lzma_stream_encoder(
+ &strm, filters, check);
+ break;
+
+ case FORMAT_LZMA:
+ ret = lzma_alone_encoder(&strm, filters[0].options);
+ break;
+
+# ifdef HAVE_LZIP_DECODER
+ case FORMAT_LZIP:
+ // args.c should disallow this.
+ assert(0);
+ ret = LZMA_PROG_ERROR;
+ break;
+# endif
+
+ case FORMAT_RAW:
+ ret = lzma_raw_encoder(&strm, filters);
+ break;
+ }
+#endif
+ } else {
+#ifdef HAVE_DECODERS
+ uint32_t flags = 0;
+
+ // It seems silly to warn about unsupported check if the
+ // check won't be verified anyway due to --ignore-check.
+ if (opt_ignore_check)
+ flags |= LZMA_IGNORE_CHECK;
+ else
+ flags |= LZMA_TELL_UNSUPPORTED_CHECK;
+
+ if (opt_single_stream)
+ allow_trailing_input = true;
+ else
+ flags |= LZMA_CONCATENATED;
+
+ // We abuse FORMAT_AUTO to indicate unknown file format,
+ // for which we may consider passthru mode.
+ enum format_type init_format = FORMAT_AUTO;
+
+ switch (opt_format) {
+ case FORMAT_AUTO:
+ // .lz is checked before .lzma since .lzma detection
+ // is more complicated (no magic bytes).
+ if (is_format_xz())
+ init_format = FORMAT_XZ;
+# ifdef HAVE_LZIP_DECODER
+ else if (is_format_lzip())
+ init_format = FORMAT_LZIP;
+# endif
+ else if (is_format_lzma())
+ init_format = FORMAT_LZMA;
+ break;
+
+ case FORMAT_XZ:
+ if (is_format_xz())
+ init_format = FORMAT_XZ;
+ break;
+
+ case FORMAT_LZMA:
+ if (is_format_lzma())
+ init_format = FORMAT_LZMA;
+ break;
+
+# ifdef HAVE_LZIP_DECODER
+ case FORMAT_LZIP:
+ if (is_format_lzip())
+ init_format = FORMAT_LZIP;
+ break;
+# endif
+
+ case FORMAT_RAW:
+ init_format = FORMAT_RAW;
+ break;
+ }
+
+ switch (init_format) {
+ case FORMAT_AUTO:
+ // Unknown file format. If --decompress --stdout
+ // --force have been given, then we copy the input
+ // as is to stdout. Checking for MODE_DECOMPRESS
+ // is needed, because we don't want to do use
+ // passthru mode with --test.
+ if (opt_mode == MODE_DECOMPRESS
+ && opt_stdout && opt_force) {
+ // These are needed for progress info.
+ strm.total_in = 0;
+ strm.total_out = 0;
+ return CODER_INIT_PASSTHRU;
+ }
+
+ ret = LZMA_FORMAT_ERROR;
+ break;
+
+ case FORMAT_XZ:
+# ifdef MYTHREAD_ENABLED
+ mt_options.flags = flags;
+
+ mt_options.threads = hardware_threads_get();
+ mt_options.memlimit_stop
+ = hardware_memlimit_get(MODE_DECOMPRESS);
+
+ // If single-threaded mode was requested, set the
+ // memlimit for threading to zero. This forces the
+ // decoder to use single-threaded mode which matches
+ // the behavior of lzma_stream_decoder().
+ //
+ // Otherwise use the limit for threaded decompression
+ // which has a sane default (users are still free to
+ // make it insanely high though).
+ mt_options.memlimit_threading
+ = mt_options.threads == 1
+ ? 0 : hardware_memlimit_mtdec_get();
+
+ ret = lzma_stream_decoder_mt(&strm, &mt_options);
+# else
+ ret = lzma_stream_decoder(&strm,
+ hardware_memlimit_get(
+ MODE_DECOMPRESS), flags);
+# endif
+ break;
+
+ case FORMAT_LZMA:
+ ret = lzma_alone_decoder(&strm,
+ hardware_memlimit_get(
+ MODE_DECOMPRESS));
+ break;
+
+# ifdef HAVE_LZIP_DECODER
+ case FORMAT_LZIP:
+ allow_trailing_input = true;
+ ret = lzma_lzip_decoder(&strm,
+ hardware_memlimit_get(
+ MODE_DECOMPRESS), flags);
+ break;
+# endif
+
+ case FORMAT_RAW:
+ // Memory usage has already been checked in
+ // coder_set_compression_settings().
+ ret = lzma_raw_decoder(&strm, filters);
+ break;
+ }
+
+ // Try to decode the headers. This will catch too low
+ // memory usage limit in case it happens in the first
+ // Block of the first Stream, which is where it very
+ // probably will happen if it is going to happen.
+ //
+ // This will also catch unsupported check type which
+ // we treat as a warning only. If there are empty
+ // concatenated Streams with unsupported check type then
+ // the message can be shown more than once here. The loop
+ // is used in case there is first a warning about
+ // unsupported check type and then the first Block
+ // would exceed the memlimit.
+ if (ret == LZMA_OK && init_format != FORMAT_RAW) {
+ strm.next_out = NULL;
+ strm.avail_out = 0;
+ while ((ret = lzma_code(&strm, LZMA_RUN))
+ == LZMA_UNSUPPORTED_CHECK)
+ message_warning(_("%s: %s"), pair->src_name,
+ message_strm(ret));
+
+ // With --single-stream lzma_code won't wait for
+ // LZMA_FINISH and thus it can return LZMA_STREAM_END
+ // if the file has no uncompressed data inside.
+ // So treat LZMA_STREAM_END as LZMA_OK here.
+ // When lzma_code() is called again in coder_normal()
+ // it will return LZMA_STREAM_END again.
+ if (ret == LZMA_STREAM_END)
+ ret = LZMA_OK;
+ }
+#endif
+ }
+
+ if (ret != LZMA_OK) {
+ message_error(_("%s: %s"), pair->src_name, message_strm(ret));
+ if (ret == LZMA_MEMLIMIT_ERROR)
+ message_mem_needed(V_ERROR, lzma_memusage(&strm));
+
+ return CODER_INIT_ERROR;
+ }
+
+ return CODER_INIT_NORMAL;
+}
+
+
+/// Resolve conflicts between opt_block_size and opt_block_list in single
+/// threaded mode. We want to default to opt_block_list, except when it is
+/// larger than opt_block_size. If this is the case for the current Block
+/// at *list_pos, then we break into smaller Blocks. Otherwise advance
+/// to the next Block in opt_block_list, and break apart if needed.
+static void
+split_block(uint64_t *block_remaining,
+ uint64_t *next_block_remaining,
+ size_t *list_pos)
+{
+ if (*next_block_remaining > 0) {
+ // The Block at *list_pos has previously been split up.
+ assert(!hardware_threads_is_mt());
+ assert(opt_block_size > 0);
+ assert(opt_block_list != NULL);
+
+ if (*next_block_remaining > opt_block_size) {
+ // We have to split the current Block at *list_pos
+ // into another opt_block_size length Block.
+ *block_remaining = opt_block_size;
+ } else {
+ // This is the last remaining split Block for the
+ // Block at *list_pos.
+ *block_remaining = *next_block_remaining;
+ }
+
+ *next_block_remaining -= *block_remaining;
+
+ } else {
+ // The Block at *list_pos has been finished. Go to the next
+ // entry in the list. If the end of the list has been reached,
+ // reuse the size of the last Block.
+ if (opt_block_list[*list_pos + 1] != 0)
+ ++*list_pos;
+
+ *block_remaining = opt_block_list[*list_pos];
+
+ // If in single-threaded mode, split up the Block if needed.
+ // This is not needed in multi-threaded mode because liblzma
+ // will do this due to how threaded encoding works.
+ if (!hardware_threads_is_mt() && opt_block_size > 0
+ && *block_remaining > opt_block_size) {
+ *next_block_remaining
+ = *block_remaining - opt_block_size;
+ *block_remaining = opt_block_size;
+ }
+ }
+}
+
+
+static bool
+coder_write_output(file_pair *pair)
+{
+ if (opt_mode != MODE_TEST) {
+ if (io_write(pair, &out_buf, IO_BUFFER_SIZE - strm.avail_out))
+ return true;
+ }
+
+ strm.next_out = out_buf.u8;
+ strm.avail_out = IO_BUFFER_SIZE;
+ return false;
+}
+
+
+/// Compress or decompress using liblzma.
+static bool
+coder_normal(file_pair *pair)
+{
+ // Encoder needs to know when we have given all the input to it.
+ // The decoders need to know it too when we are using
+ // LZMA_CONCATENATED. We need to check for src_eof here, because
+ // the first input chunk has been already read if decompressing,
+ // and that may have been the only chunk we will read.
+ lzma_action action = pair->src_eof ? LZMA_FINISH : LZMA_RUN;
+
+ lzma_ret ret;
+
+ // Assume that something goes wrong.
+ bool success = false;
+
+ // block_remaining indicates how many input bytes to encode before
+ // finishing the current .xz Block. The Block size is set with
+ // --block-size=SIZE and --block-list. They have an effect only when
+ // compressing to the .xz format. If block_remaining == UINT64_MAX,
+ // only a single block is created.
+ uint64_t block_remaining = UINT64_MAX;
+
+ // next_block_remaining for when we are in single-threaded mode and
+ // the Block in --block-list is larger than the --block-size=SIZE.
+ uint64_t next_block_remaining = 0;
+
+ // Position in opt_block_list. Unused if --block-list wasn't used.
+ size_t list_pos = 0;
+
+ // Handle --block-size for single-threaded mode and the first step
+ // of --block-list.
+ if (opt_mode == MODE_COMPRESS && opt_format == FORMAT_XZ) {
+ // --block-size doesn't do anything here in threaded mode,
+ // because the threaded encoder will take care of splitting
+ // to fixed-sized Blocks.
+ if (!hardware_threads_is_mt() && opt_block_size > 0)
+ block_remaining = opt_block_size;
+
+ // If --block-list was used, start with the first size.
+ //
+ // For threaded case, --block-size specifies how big Blocks
+ // the encoder needs to be prepared to create at maximum
+ // and --block-list will simultaneously cause new Blocks
+ // to be started at specified intervals. To keep things
+ // logical, the same is done in single-threaded mode. The
+ // output is still not identical because in single-threaded
+ // mode the size info isn't written into Block Headers.
+ if (opt_block_list != NULL) {
+ if (block_remaining < opt_block_list[list_pos]) {
+ assert(!hardware_threads_is_mt());
+ next_block_remaining = opt_block_list[list_pos]
+ - block_remaining;
+ } else {
+ block_remaining = opt_block_list[list_pos];
+ }
+ }
+ }
+
+ strm.next_out = out_buf.u8;
+ strm.avail_out = IO_BUFFER_SIZE;
+
+ while (!user_abort) {
+ // Fill the input buffer if it is empty and we aren't
+ // flushing or finishing.
+ if (strm.avail_in == 0 && action == LZMA_RUN) {
+ strm.next_in = in_buf.u8;
+ strm.avail_in = io_read(pair, &in_buf,
+ my_min(block_remaining,
+ IO_BUFFER_SIZE));
+
+ if (strm.avail_in == SIZE_MAX)
+ break;
+
+ if (pair->src_eof) {
+ action = LZMA_FINISH;
+
+ } else if (block_remaining != UINT64_MAX) {
+ // Start a new Block after every
+ // opt_block_size bytes of input.
+ block_remaining -= strm.avail_in;
+ if (block_remaining == 0)
+ action = LZMA_FULL_BARRIER;
+ }
+
+ if (action == LZMA_RUN && pair->flush_needed)
+ action = LZMA_SYNC_FLUSH;
+ }
+
+ // Let liblzma do the actual work.
+ ret = lzma_code(&strm, action);
+
+ // Write out if the output buffer became full.
+ if (strm.avail_out == 0) {
+ if (coder_write_output(pair))
+ break;
+ }
+
+ if (ret == LZMA_STREAM_END && (action == LZMA_SYNC_FLUSH
+ || action == LZMA_FULL_BARRIER)) {
+ if (action == LZMA_SYNC_FLUSH) {
+ // Flushing completed. Write the pending data
+ // out immediately so that the reading side
+ // can decompress everything compressed so far.
+ if (coder_write_output(pair))
+ break;
+
+ // Mark that we haven't seen any new input
+ // since the previous flush.
+ pair->src_has_seen_input = false;
+ pair->flush_needed = false;
+ } else {
+ // Start a new Block after LZMA_FULL_BARRIER.
+ if (opt_block_list == NULL) {
+ assert(!hardware_threads_is_mt());
+ assert(opt_block_size > 0);
+ block_remaining = opt_block_size;
+ } else {
+ split_block(&block_remaining,
+ &next_block_remaining,
+ &list_pos);
+ }
+ }
+
+ // Start a new Block after LZMA_FULL_FLUSH or continue
+ // the same block after LZMA_SYNC_FLUSH.
+ action = LZMA_RUN;
+
+ } else if (ret != LZMA_OK) {
+ // Determine if the return value indicates that we
+ // won't continue coding. LZMA_NO_CHECK would be
+ // here too if LZMA_TELL_ANY_CHECK was used.
+ const bool stop = ret != LZMA_UNSUPPORTED_CHECK;
+
+ if (stop) {
+ // Write the remaining bytes even if something
+ // went wrong, because that way the user gets
+ // as much data as possible, which can be good
+ // when trying to get at least some useful
+ // data out of damaged files.
+ if (coder_write_output(pair))
+ break;
+ }
+
+ if (ret == LZMA_STREAM_END) {
+ if (allow_trailing_input) {
+ io_fix_src_pos(pair, strm.avail_in);
+ success = true;
+ break;
+ }
+
+ // Check that there is no trailing garbage.
+ // This is needed for LZMA_Alone and raw
+ // streams. This is *not* done with .lz files
+ // as that format specifically requires
+ // allowing trailing garbage.
+ if (strm.avail_in == 0 && !pair->src_eof) {
+ // Try reading one more byte.
+ // Hopefully we don't get any more
+ // input, and thus pair->src_eof
+ // becomes true.
+ strm.avail_in = io_read(
+ pair, &in_buf, 1);
+ if (strm.avail_in == SIZE_MAX)
+ break;
+
+ assert(strm.avail_in == 0
+ || strm.avail_in == 1);
+ }
+
+ if (strm.avail_in == 0) {
+ assert(pair->src_eof);
+ success = true;
+ break;
+ }
+
+ // We hadn't reached the end of the file.
+ ret = LZMA_DATA_ERROR;
+ assert(stop);
+ }
+
+ // If we get here and stop is true, something went
+ // wrong and we print an error. Otherwise it's just
+ // a warning and coding can continue.
+ if (stop) {
+ message_error(_("%s: %s"), pair->src_name,
+ message_strm(ret));
+ } else {
+ message_warning(_("%s: %s"), pair->src_name,
+ message_strm(ret));
+
+ // When compressing, all possible errors set
+ // stop to true.
+ assert(opt_mode != MODE_COMPRESS);
+ }
+
+ if (ret == LZMA_MEMLIMIT_ERROR) {
+ // Display how much memory it would have
+ // actually needed.
+ message_mem_needed(V_ERROR,
+ lzma_memusage(&strm));
+ }
+
+ if (stop)
+ break;
+ }
+
+ // Show progress information under certain conditions.
+ message_progress_update();
+ }
+
+ return success;
+}
+
+
+/// Copy from input file to output file without processing the data in any
+/// way. This is used only when trying to decompress unrecognized files
+/// with --decompress --stdout --force, so the output is always stdout.
+static bool
+coder_passthru(file_pair *pair)
+{
+ while (strm.avail_in != 0) {
+ if (user_abort)
+ return false;
+
+ if (io_write(pair, &in_buf, strm.avail_in))
+ return false;
+
+ strm.total_in += strm.avail_in;
+ strm.total_out = strm.total_in;
+ message_progress_update();
+
+ strm.avail_in = io_read(pair, &in_buf, IO_BUFFER_SIZE);
+ if (strm.avail_in == SIZE_MAX)
+ return false;
+ }
+
+ return true;
+}
+
+
+extern void
+coder_run(const char *filename)
+{
+ // Set and possibly print the filename for the progress message.
+ message_filename(filename);
+
+ // Try to open the input file.
+ file_pair *pair = io_open_src(filename);
+ if (pair == NULL)
+ return;
+
+ // Assume that something goes wrong.
+ bool success = false;
+
+ if (opt_mode == MODE_COMPRESS) {
+ strm.next_in = NULL;
+ strm.avail_in = 0;
+ } else {
+ // Read the first chunk of input data. This is needed
+ // to detect the input file type.
+ strm.next_in = in_buf.u8;
+ strm.avail_in = io_read(pair, &in_buf, IO_BUFFER_SIZE);
+ }
+
+ if (strm.avail_in != SIZE_MAX) {
+ // Initialize the coder. This will detect the file format
+ // and, in decompression or testing mode, check the memory
+ // usage of the first Block too. This way we don't try to
+ // open the destination file if we see that coding wouldn't
+ // work at all anyway. This also avoids deleting the old
+ // "target" file if --force was used.
+ const enum coder_init_ret init_ret = coder_init(pair);
+
+ if (init_ret != CODER_INIT_ERROR && !user_abort) {
+ // Don't open the destination file when --test
+ // is used.
+ if (opt_mode == MODE_TEST || !io_open_dest(pair)) {
+ // Remember the current time. It is needed
+ // for progress indicator.
+ mytime_set_start_time();
+
+ // Initialize the progress indicator.
+ //
+ // NOTE: When reading from stdin, fstat()
+ // isn't called on it and thus src_st.st_size
+ // is zero. If stdin pointed to a regular
+ // file, it would still be possible to know
+ // the file size but then we would also need
+ // to take into account the current reading
+ // position since with stdin it isn't
+ // necessarily at the beginning of the file.
+ const bool is_passthru = init_ret
+ == CODER_INIT_PASSTHRU;
+ const uint64_t in_size
+ = pair->src_st.st_size <= 0
+ ? 0 : (uint64_t)(pair->src_st.st_size);
+ message_progress_start(&strm,
+ is_passthru, in_size);
+
+ // Do the actual coding or passthru.
+ if (is_passthru)
+ success = coder_passthru(pair);
+ else
+ success = coder_normal(pair);
+
+ message_progress_end(success);
+ }
+ }
+ }
+
+ // Close the file pair. It needs to know if coding was successful to
+ // know if the source or target file should be unlinked.
+ io_close(pair, success);
+
+ return;
+}
+
+
+#ifndef NDEBUG
+extern void
+coder_free(void)
+{
+ lzma_end(&strm);
+ return;
+}
+#endif
diff --git a/src/xz/coder.h b/src/xz/coder.h
new file mode 100644
index 0000000..b4f43a2
--- /dev/null
+++ b/src/xz/coder.h
@@ -0,0 +1,79 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file coder.h
+/// \brief Compresses or uncompresses a file
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+enum operation_mode {
+ MODE_COMPRESS,
+ MODE_DECOMPRESS,
+ MODE_TEST,
+ MODE_LIST,
+};
+
+
+// NOTE: The order of these is significant in suffix.c.
+enum format_type {
+ FORMAT_AUTO,
+ FORMAT_XZ,
+ FORMAT_LZMA,
+#ifdef HAVE_LZIP_DECODER
+ FORMAT_LZIP,
+#endif
+ FORMAT_RAW,
+};
+
+
+/// Operation mode of the command line tool. This is set in args.c and read
+/// in several files.
+extern enum operation_mode opt_mode;
+
+/// File format to use when encoding or what format(s) to accept when
+/// decoding. This is a global because it's needed also in suffix.c.
+/// This is set in args.c.
+extern enum format_type opt_format;
+
+/// If true, the compression settings are automatically adjusted down if
+/// they exceed the memory usage limit.
+extern bool opt_auto_adjust;
+
+/// If true, stop after decoding the first stream.
+extern bool opt_single_stream;
+
+/// If non-zero, start a new .xz Block after every opt_block_size bytes
+/// of input. This has an effect only when compressing to the .xz format.
+extern uint64_t opt_block_size;
+
+/// This is non-NULL if --block-list was used. This contains the Block sizes
+/// as an array that is terminated with 0.
+extern uint64_t *opt_block_list;
+
+/// Set the integrity check type used when compressing
+extern void coder_set_check(lzma_check check);
+
+/// Set preset number
+extern void coder_set_preset(uint32_t new_preset);
+
+/// Enable extreme mode
+extern void coder_set_extreme(void);
+
+/// Add a filter to the custom filter chain
+extern void coder_add_filter(lzma_vli id, void *options);
+
+/// Set and partially validate compression settings. This can also be used
+/// in decompression or test mode with the raw format.
+extern void coder_set_compression_settings(void);
+
+/// Compress or decompress the given file
+extern void coder_run(const char *filename);
+
+#ifndef NDEBUG
+/// Free the memory allocated for the coder and kill the worker threads.
+extern void coder_free(void);
+#endif
diff --git a/src/xz/file_io.c b/src/xz/file_io.c
new file mode 100644
index 0000000..2828029
--- /dev/null
+++ b/src/xz/file_io.c
@@ -0,0 +1,1403 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file file_io.c
+/// \brief File opening, unlinking, and closing
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+#include <fcntl.h>
+
+#ifdef TUKLIB_DOSLIKE
+# include <io.h>
+#else
+# include <poll.h>
+static bool warn_fchown;
+#endif
+
+#if defined(HAVE_FUTIMES) || defined(HAVE_FUTIMESAT) || defined(HAVE_UTIMES)
+# include <sys/time.h>
+#elif defined(HAVE__FUTIME)
+# include <sys/utime.h>
+#elif defined(HAVE_UTIME)
+# include <utime.h>
+#endif
+
+#ifdef HAVE_CAPSICUM
+# ifdef HAVE_SYS_CAPSICUM_H
+# include <sys/capsicum.h>
+# else
+# include <sys/capability.h>
+# endif
+#endif
+
+#include "tuklib_open_stdxxx.h"
+
+#ifndef O_BINARY
+# define O_BINARY 0
+#endif
+
+#ifndef O_NOCTTY
+# define O_NOCTTY 0
+#endif
+
+// Using this macro to silence a warning from gcc -Wlogical-op.
+#if EAGAIN == EWOULDBLOCK
+# define IS_EAGAIN_OR_EWOULDBLOCK(e) ((e) == EAGAIN)
+#else
+# define IS_EAGAIN_OR_EWOULDBLOCK(e) \
+ ((e) == EAGAIN || (e) == EWOULDBLOCK)
+#endif
+
+
+typedef enum {
+ IO_WAIT_MORE, // Reading or writing is possible.
+ IO_WAIT_ERROR, // Error or user_abort
+ IO_WAIT_TIMEOUT, // poll() timed out
+} io_wait_ret;
+
+
+/// If true, try to create sparse files when decompressing.
+static bool try_sparse = true;
+
+#ifdef ENABLE_SANDBOX
+/// True if the conditions for sandboxing (described in main()) have been met.
+static bool sandbox_allowed = false;
+#endif
+
+#ifndef TUKLIB_DOSLIKE
+/// File status flags of standard input. This is used by io_open_src()
+/// and io_close_src().
+static int stdin_flags;
+static bool restore_stdin_flags = false;
+
+/// Original file status flags of standard output. This is used by
+/// io_open_dest() and io_close_dest() to save and restore the flags.
+static int stdout_flags;
+static bool restore_stdout_flags = false;
+
+/// Self-pipe used together with the user_abort variable to avoid
+/// race conditions with signal handling.
+static int user_abort_pipe[2];
+#endif
+
+
+static bool io_write_buf(file_pair *pair, const uint8_t *buf, size_t size);
+
+
+extern void
+io_init(void)
+{
+ // Make sure that stdin, stdout, and stderr are connected to
+ // a valid file descriptor. Exit immediately with exit code ERROR
+ // if we cannot make the file descriptors valid. Maybe we should
+ // print an error message, but our stderr could be screwed anyway.
+ tuklib_open_stdxxx(E_ERROR);
+
+#ifndef TUKLIB_DOSLIKE
+ // If fchown() fails setting the owner, we warn about it only if
+ // we are root.
+ warn_fchown = geteuid() == 0;
+
+ // Create a pipe for the self-pipe trick.
+ if (pipe(user_abort_pipe))
+ message_fatal(_("Error creating a pipe: %s"),
+ strerror(errno));
+
+ // Make both ends of the pipe non-blocking.
+ for (unsigned i = 0; i < 2; ++i) {
+ int flags = fcntl(user_abort_pipe[i], F_GETFL);
+ if (flags == -1 || fcntl(user_abort_pipe[i], F_SETFL,
+ flags | O_NONBLOCK) == -1)
+ message_fatal(_("Error creating a pipe: %s"),
+ strerror(errno));
+ }
+#endif
+
+#ifdef __DJGPP__
+ // Avoid doing useless things when statting files.
+ // This isn't important but doesn't hurt.
+ _djstat_flags = _STAT_EXEC_EXT | _STAT_EXEC_MAGIC | _STAT_DIRSIZE;
+#endif
+
+ return;
+}
+
+
+#ifndef TUKLIB_DOSLIKE
+extern void
+io_write_to_user_abort_pipe(void)
+{
+ // If the write() fails, it's probably due to the pipe being full.
+ // Failing in that case is fine. If the reason is something else,
+ // there's not much we can do since this is called in a signal
+ // handler. So ignore the errors and try to avoid warnings with
+ // GCC and glibc when _FORTIFY_SOURCE=2 is used.
+ uint8_t b = '\0';
+ const ssize_t ret = write(user_abort_pipe[1], &b, 1);
+ (void)ret;
+ return;
+}
+#endif
+
+
+extern void
+io_no_sparse(void)
+{
+ try_sparse = false;
+ return;
+}
+
+
+#ifdef ENABLE_SANDBOX
+extern void
+io_allow_sandbox(void)
+{
+ sandbox_allowed = true;
+ return;
+}
+
+
+/// Enables operating-system-specific sandbox if it is possible.
+/// src_fd is the file descriptor of the input file.
+static void
+io_sandbox_enter(int src_fd)
+{
+ if (!sandbox_allowed) {
+ // This message is more often annoying than useful so
+ // it's commented out. It can be useful when developing
+ // the sandboxing code.
+ //message(V_DEBUG, _("Sandbox is disabled due "
+ // "to incompatible command line arguments"));
+ return;
+ }
+
+ const char dummy_str[] = "x";
+
+ // Try to ensure that both libc and xz locale files have been
+ // loaded when NLS is enabled.
+ snprintf(NULL, 0, "%s%s", _(dummy_str), strerror(EINVAL));
+
+ // Try to ensure that iconv data files needed for handling multibyte
+ // characters have been loaded. This is needed at least with glibc.
+ tuklib_mbstr_width(dummy_str, NULL);
+
+#ifdef HAVE_CAPSICUM
+ // Capsicum needs FreeBSD 10.0 or later.
+ cap_rights_t rights;
+
+ if (cap_enter())
+ goto error;
+
+ if (cap_rights_limit(src_fd, cap_rights_init(&rights,
+ CAP_EVENT, CAP_FCNTL, CAP_LOOKUP, CAP_READ, CAP_SEEK)))
+ goto error;
+
+ if (src_fd != STDIN_FILENO && cap_rights_limit(
+ STDIN_FILENO, cap_rights_clear(&rights)))
+ goto error;
+
+ if (cap_rights_limit(STDOUT_FILENO, cap_rights_init(&rights,
+ CAP_EVENT, CAP_FCNTL, CAP_FSTAT, CAP_LOOKUP,
+ CAP_WRITE, CAP_SEEK)))
+ goto error;
+
+ if (cap_rights_limit(STDERR_FILENO, cap_rights_init(&rights,
+ CAP_WRITE)))
+ goto error;
+
+ if (cap_rights_limit(user_abort_pipe[0], cap_rights_init(&rights,
+ CAP_EVENT)))
+ goto error;
+
+ if (cap_rights_limit(user_abort_pipe[1], cap_rights_init(&rights,
+ CAP_WRITE)))
+ goto error;
+
+#elif defined(HAVE_PLEDGE)
+ // pledge() was introduced in OpenBSD 5.9.
+ //
+ // main() unconditionally calls pledge() with fairly relaxed
+ // promises which work in all situations. Here we make the
+ // sandbox more strict.
+ if (pledge("stdio", ""))
+ goto error;
+
+ (void)src_fd;
+
+#else
+# error ENABLE_SANDBOX is defined but no sandboxing method was found.
+#endif
+
+ // This message is annoying in xz -lvv.
+ //message(V_DEBUG, _("Sandbox was successfully enabled"));
+ return;
+
+error:
+#ifdef HAVE_CAPSICUM
+ // If a kernel is configured without capability mode support or
+ // used in an emulator that does not implement the capability
+ // system calls, then the Capsicum system calls will fail and set
+ // errno to ENOSYS. In that case xz will silently run without
+ // the sandbox.
+ if (errno == ENOSYS)
+ return;
+#endif
+ message_fatal(_("Failed to enable the sandbox"));
+}
+#endif // ENABLE_SANDBOX
+
+
+#ifndef TUKLIB_DOSLIKE
+/// \brief Waits for input or output to become available or for a signal
+///
+/// This uses the self-pipe trick to avoid a race condition that can occur
+/// if a signal is caught after user_abort has been checked but before e.g.
+/// read() has been called. In that situation read() could block unless
+/// non-blocking I/O is used. With non-blocking I/O something like select()
+/// or poll() is needed to avoid a busy-wait loop, and the same race condition
+/// pops up again. There are pselect() (POSIX-1.2001) and ppoll() (not in
+/// POSIX) but neither is portable enough in 2013. The self-pipe trick is
+/// old and very portable.
+static io_wait_ret
+io_wait(file_pair *pair, int timeout, bool is_reading)
+{
+ struct pollfd pfd[2];
+
+ if (is_reading) {
+ pfd[0].fd = pair->src_fd;
+ pfd[0].events = POLLIN;
+ } else {
+ pfd[0].fd = pair->dest_fd;
+ pfd[0].events = POLLOUT;
+ }
+
+ pfd[1].fd = user_abort_pipe[0];
+ pfd[1].events = POLLIN;
+
+ while (true) {
+ const int ret = poll(pfd, 2, timeout);
+
+ if (user_abort)
+ return IO_WAIT_ERROR;
+
+ if (ret == -1) {
+ if (errno == EINTR || errno == EAGAIN)
+ continue;
+
+ message_error(_("%s: poll() failed: %s"),
+ is_reading ? pair->src_name
+ : pair->dest_name,
+ strerror(errno));
+ return IO_WAIT_ERROR;
+ }
+
+ if (ret == 0)
+ return IO_WAIT_TIMEOUT;
+
+ if (pfd[0].revents != 0)
+ return IO_WAIT_MORE;
+ }
+}
+#endif
+
+
+/// \brief Unlink a file
+///
+/// This tries to verify that the file being unlinked really is the file that
+/// we want to unlink by verifying device and inode numbers. There's still
+/// a small unavoidable race, but this is much better than nothing (the file
+/// could have been moved/replaced even hours earlier).
+static void
+io_unlink(const char *name, const struct stat *known_st)
+{
+#if defined(TUKLIB_DOSLIKE)
+ // On DOS-like systems, st_ino is meaningless, so don't bother
+ // testing it. Just silence a compiler warning.
+ (void)known_st;
+#else
+ struct stat new_st;
+
+ // If --force was used, use stat() instead of lstat(). This way
+ // (de)compressing symlinks works correctly. However, it also means
+ // that xz cannot detect if a regular file foo is renamed to bar
+ // and then a symlink foo -> bar is created. Because of stat()
+ // instead of lstat(), xz will think that foo hasn't been replaced
+ // with another file. Thus, xz will remove foo even though it no
+ // longer is the same file that xz used when it started compressing.
+ // Probably it's not too bad though, so this doesn't need a more
+ // complex fix.
+ const int stat_ret = opt_force
+ ? stat(name, &new_st) : lstat(name, &new_st);
+
+ if (stat_ret
+# ifdef __VMS
+ // st_ino is an array, and we don't want to
+ // compare st_dev at all.
+ || memcmp(&new_st.st_ino, &known_st->st_ino,
+ sizeof(new_st.st_ino)) != 0
+# else
+ // Typical POSIX-like system
+ || new_st.st_dev != known_st->st_dev
+ || new_st.st_ino != known_st->st_ino
+# endif
+ )
+ // TRANSLATORS: When compression or decompression finishes,
+ // and xz is going to remove the source file, xz first checks
+ // if the source file still exists, and if it does, does its
+ // device and inode numbers match what xz saw when it opened
+ // the source file. If these checks fail, this message is
+ // shown, %s being the filename, and the file is not deleted.
+ // The check for device and inode numbers is there, because
+ // it is possible that the user has put a new file in place
+ // of the original file, and in that case it obviously
+ // shouldn't be removed.
+ message_warning(_("%s: File seems to have been moved, "
+ "not removing"), name);
+ else
+#endif
+ // There's a race condition between lstat() and unlink()
+ // but at least we have tried to avoid removing wrong file.
+ if (unlink(name))
+ message_warning(_("%s: Cannot remove: %s"),
+ name, strerror(errno));
+
+ return;
+}
+
+
+/// \brief Copies owner/group and permissions
+///
+/// \todo ACL and EA support
+///
+static void
+io_copy_attrs(const file_pair *pair)
+{
+ // Skip chown and chmod on Windows.
+#ifndef TUKLIB_DOSLIKE
+ // This function is more tricky than you may think at first.
+ // Blindly copying permissions may permit users to access the
+ // destination file who didn't have permission to access the
+ // source file.
+
+ // Try changing the owner of the file. If we aren't root or the owner
+ // isn't already us, fchown() probably doesn't succeed. We warn
+ // about failing fchown() only if we are root.
+ if (fchown(pair->dest_fd, pair->src_st.st_uid, (gid_t)(-1))
+ && warn_fchown)
+ message_warning(_("%s: Cannot set the file owner: %s"),
+ pair->dest_name, strerror(errno));
+
+ mode_t mode;
+
+ // With BSD semantics the new dest file may have a group that
+ // does not belong to the user. If the src file has the same gid
+ // nothing has to be done. Nevertheless OpenBSD fchown(2) fails
+ // in this case which seems to be POSIX compliant. As there is
+ // nothing to do, skip the system call.
+ if (pair->dest_st.st_gid != pair->src_st.st_gid
+ && fchown(pair->dest_fd, (uid_t)(-1),
+ pair->src_st.st_gid)) {
+ message_warning(_("%s: Cannot set the file group: %s"),
+ pair->dest_name, strerror(errno));
+ // We can still safely copy some additional permissions:
+ // `group' must be at least as strict as `other' and
+ // also vice versa.
+ //
+ // NOTE: After this, the owner of the source file may
+ // get additional permissions. This shouldn't be too bad,
+ // because the owner would have had permission to chmod
+ // the original file anyway.
+ mode = ((pair->src_st.st_mode & 0070) >> 3)
+ & (pair->src_st.st_mode & 0007);
+ mode = (pair->src_st.st_mode & 0700) | (mode << 3) | mode;
+ } else {
+ // Drop the setuid, setgid, and sticky bits.
+ mode = pair->src_st.st_mode & 0777;
+ }
+
+ if (fchmod(pair->dest_fd, mode))
+ message_warning(_("%s: Cannot set the file permissions: %s"),
+ pair->dest_name, strerror(errno));
+#endif
+
+ // Copy the timestamps. We have several possible ways to do this, of
+ // which some are better in both security and precision.
+ //
+ // First, get the nanosecond part of the timestamps. As of writing,
+ // it's not standardized by POSIX, and there are several names for
+ // the same thing in struct stat.
+ long atime_nsec;
+ long mtime_nsec;
+
+# if defined(HAVE_STRUCT_STAT_ST_ATIM_TV_NSEC)
+ // GNU and Solaris
+ atime_nsec = pair->src_st.st_atim.tv_nsec;
+ mtime_nsec = pair->src_st.st_mtim.tv_nsec;
+
+# elif defined(HAVE_STRUCT_STAT_ST_ATIMESPEC_TV_NSEC)
+ // BSD
+ atime_nsec = pair->src_st.st_atimespec.tv_nsec;
+ mtime_nsec = pair->src_st.st_mtimespec.tv_nsec;
+
+# elif defined(HAVE_STRUCT_STAT_ST_ATIMENSEC)
+ // GNU and BSD without extensions
+ atime_nsec = pair->src_st.st_atimensec;
+ mtime_nsec = pair->src_st.st_mtimensec;
+
+# elif defined(HAVE_STRUCT_STAT_ST_UATIME)
+ // Tru64
+ atime_nsec = pair->src_st.st_uatime * 1000;
+ mtime_nsec = pair->src_st.st_umtime * 1000;
+
+# elif defined(HAVE_STRUCT_STAT_ST_ATIM_ST__TIM_TV_NSEC)
+ // UnixWare
+ atime_nsec = pair->src_st.st_atim.st__tim.tv_nsec;
+ mtime_nsec = pair->src_st.st_mtim.st__tim.tv_nsec;
+
+# else
+ // Safe fallback
+ atime_nsec = 0;
+ mtime_nsec = 0;
+# endif
+
+ // Construct a structure to hold the timestamps and call appropriate
+ // function to set the timestamps.
+#if defined(HAVE_FUTIMENS)
+ // Use nanosecond precision.
+ struct timespec tv[2];
+ tv[0].tv_sec = pair->src_st.st_atime;
+ tv[0].tv_nsec = atime_nsec;
+ tv[1].tv_sec = pair->src_st.st_mtime;
+ tv[1].tv_nsec = mtime_nsec;
+
+ (void)futimens(pair->dest_fd, tv);
+
+#elif defined(HAVE_FUTIMES) || defined(HAVE_FUTIMESAT) || defined(HAVE_UTIMES)
+ // Use microsecond precision.
+ struct timeval tv[2];
+ tv[0].tv_sec = pair->src_st.st_atime;
+ tv[0].tv_usec = atime_nsec / 1000;
+ tv[1].tv_sec = pair->src_st.st_mtime;
+ tv[1].tv_usec = mtime_nsec / 1000;
+
+# if defined(HAVE_FUTIMES)
+ (void)futimes(pair->dest_fd, tv);
+# elif defined(HAVE_FUTIMESAT)
+ (void)futimesat(pair->dest_fd, NULL, tv);
+# else
+ // Argh, no function to use a file descriptor to set the timestamp.
+ (void)utimes(pair->dest_name, tv);
+# endif
+
+#elif defined(HAVE__FUTIME)
+ // Use one-second precision with Windows-specific _futime().
+ // We could use utime() too except that for some reason the
+ // timestamp will get reset at close(). With _futime() it works.
+ // This struct cannot be const as _futime() takes a non-const pointer.
+ struct _utimbuf buf = {
+ .actime = pair->src_st.st_atime,
+ .modtime = pair->src_st.st_mtime,
+ };
+
+ // Avoid warnings.
+ (void)atime_nsec;
+ (void)mtime_nsec;
+
+ (void)_futime(pair->dest_fd, &buf);
+
+#elif defined(HAVE_UTIME)
+ // Use one-second precision. utime() doesn't support using file
+ // descriptor either. Some systems have broken utime() prototype
+ // so don't make this const.
+ struct utimbuf buf = {
+ .actime = pair->src_st.st_atime,
+ .modtime = pair->src_st.st_mtime,
+ };
+
+ // Avoid warnings.
+ (void)atime_nsec;
+ (void)mtime_nsec;
+
+ (void)utime(pair->dest_name, &buf);
+#endif
+
+ return;
+}
+
+
+/// Opens the source file. Returns false on success, true on error.
+static bool
+io_open_src_real(file_pair *pair)
+{
+ // There's nothing to open when reading from stdin.
+ if (pair->src_name == stdin_filename) {
+ pair->src_fd = STDIN_FILENO;
+#ifdef TUKLIB_DOSLIKE
+ setmode(STDIN_FILENO, O_BINARY);
+#else
+ // Try to set stdin to non-blocking mode. It won't work
+ // e.g. on OpenBSD if stdout is e.g. /dev/null. In such
+ // case we proceed as if stdin were non-blocking anyway
+ // (in case of /dev/null it will be in practice). The
+ // same applies to stdout in io_open_dest_real().
+ stdin_flags = fcntl(STDIN_FILENO, F_GETFL);
+ if (stdin_flags == -1) {
+ message_error(_("Error getting the file status flags "
+ "from standard input: %s"),
+ strerror(errno));
+ return true;
+ }
+
+ if ((stdin_flags & O_NONBLOCK) == 0
+ && fcntl(STDIN_FILENO, F_SETFL,
+ stdin_flags | O_NONBLOCK) != -1)
+ restore_stdin_flags = true;
+#endif
+#ifdef HAVE_POSIX_FADVISE
+ // It will fail if stdin is a pipe and that's fine.
+ (void)posix_fadvise(STDIN_FILENO, 0, 0,
+ opt_mode == MODE_LIST
+ ? POSIX_FADV_RANDOM
+ : POSIX_FADV_SEQUENTIAL);
+#endif
+ return false;
+ }
+
+ // Symlinks are not followed unless writing to stdout or --force
+ // or --keep was used.
+ const bool follow_symlinks
+ = opt_stdout || opt_force || opt_keep_original;
+
+ // We accept only regular files if we are writing the output
+ // to disk too. bzip2 allows overriding this with --force but
+ // gzip and xz don't.
+ const bool reg_files_only = !opt_stdout;
+
+ // Flags for open()
+ int flags = O_RDONLY | O_BINARY | O_NOCTTY;
+
+#ifndef TUKLIB_DOSLIKE
+ // Use non-blocking I/O:
+ // - It prevents blocking when opening FIFOs and some other
+ // special files, which is good if we want to accept only
+ // regular files.
+ // - It can help avoiding some race conditions with signal handling.
+ flags |= O_NONBLOCK;
+#endif
+
+#if defined(O_NOFOLLOW)
+ if (!follow_symlinks)
+ flags |= O_NOFOLLOW;
+#elif !defined(TUKLIB_DOSLIKE)
+ // Some POSIX-like systems lack O_NOFOLLOW (it's not required
+ // by POSIX). Check for symlinks with a separate lstat() on
+ // these systems.
+ if (!follow_symlinks) {
+ struct stat st;
+ if (lstat(pair->src_name, &st)) {
+ message_error(_("%s: %s"), pair->src_name,
+ strerror(errno));
+ return true;
+
+ } else if (S_ISLNK(st.st_mode)) {
+ message_warning(_("%s: Is a symbolic link, "
+ "skipping"), pair->src_name);
+ return true;
+ }
+ }
+#else
+ // Avoid warnings.
+ (void)follow_symlinks;
+#endif
+
+ // Try to open the file. Signals have been blocked so EINTR shouldn't
+ // be possible.
+ pair->src_fd = open(pair->src_name, flags);
+
+ if (pair->src_fd == -1) {
+ // Signals (that have a signal handler) have been blocked.
+ assert(errno != EINTR);
+
+#ifdef O_NOFOLLOW
+ // Give an understandable error message if the reason
+ // for failing was that the file was a symbolic link.
+ //
+ // Note that at least Linux, OpenBSD, Solaris, and Darwin
+ // use ELOOP to indicate that O_NOFOLLOW was the reason
+ // that open() failed. Because there may be
+ // directories in the pathname, ELOOP may occur also
+ // because of a symlink loop in the directory part.
+ // So ELOOP doesn't tell us what actually went wrong,
+ // and this stupidity went into POSIX-1.2008 too.
+ //
+ // FreeBSD associates EMLINK with O_NOFOLLOW and
+ // Tru64 uses ENOTSUP. We use these directly here
+ // and skip the lstat() call and the associated race.
+ // I want to hear if there are other kernels that
+ // fail with something else than ELOOP with O_NOFOLLOW.
+ bool was_symlink = false;
+
+# if defined(__FreeBSD__) || defined(__DragonFly__)
+ if (errno == EMLINK)
+ was_symlink = true;
+
+# elif defined(__digital__) && defined(__unix__)
+ if (errno == ENOTSUP)
+ was_symlink = true;
+
+# elif defined(__NetBSD__)
+ if (errno == EFTYPE)
+ was_symlink = true;
+
+# else
+ if (errno == ELOOP && !follow_symlinks) {
+ const int saved_errno = errno;
+ struct stat st;
+ if (lstat(pair->src_name, &st) == 0
+ && S_ISLNK(st.st_mode))
+ was_symlink = true;
+
+ errno = saved_errno;
+ }
+# endif
+
+ if (was_symlink)
+ message_warning(_("%s: Is a symbolic link, "
+ "skipping"), pair->src_name);
+ else
+#endif
+ // Something else than O_NOFOLLOW failing
+ // (assuming that the race conditions didn't
+ // confuse us).
+ message_error(_("%s: %s"), pair->src_name,
+ strerror(errno));
+
+ return true;
+ }
+
+ // Stat the source file. We need the result also when we copy
+ // the permissions, and when unlinking.
+ //
+ // NOTE: Use stat() instead of fstat() with DJGPP, because
+ // then we have a better chance to get st_ino value that can
+ // be used in io_open_dest_real() to prevent overwriting the
+ // source file.
+#ifdef __DJGPP__
+ if (stat(pair->src_name, &pair->src_st))
+ goto error_msg;
+#else
+ if (fstat(pair->src_fd, &pair->src_st))
+ goto error_msg;
+#endif
+
+ if (S_ISDIR(pair->src_st.st_mode)) {
+ message_warning(_("%s: Is a directory, skipping"),
+ pair->src_name);
+ goto error;
+ }
+
+ if (reg_files_only && !S_ISREG(pair->src_st.st_mode)) {
+ message_warning(_("%s: Not a regular file, skipping"),
+ pair->src_name);
+ goto error;
+ }
+
+#ifndef TUKLIB_DOSLIKE
+ if (reg_files_only && !opt_force && !opt_keep_original) {
+ if (pair->src_st.st_mode & (S_ISUID | S_ISGID)) {
+ // gzip rejects setuid and setgid files even
+ // when --force was used. bzip2 doesn't check
+ // for them, but calls fchown() after fchmod(),
+ // and many systems automatically drop setuid
+ // and setgid bits there.
+ //
+ // We accept setuid and setgid files if
+ // --force or --keep was used. We drop these bits
+ // explicitly in io_copy_attr().
+ message_warning(_("%s: File has setuid or "
+ "setgid bit set, skipping"),
+ pair->src_name);
+ goto error;
+ }
+
+ if (pair->src_st.st_mode & S_ISVTX) {
+ message_warning(_("%s: File has sticky bit "
+ "set, skipping"),
+ pair->src_name);
+ goto error;
+ }
+
+ if (pair->src_st.st_nlink > 1) {
+ message_warning(_("%s: Input file has more "
+ "than one hard link, "
+ "skipping"), pair->src_name);
+ goto error;
+ }
+ }
+
+ // If it is something else than a regular file, wait until
+ // there is input available. This way reading from FIFOs
+ // will work when open() is used with O_NONBLOCK.
+ if (!S_ISREG(pair->src_st.st_mode)) {
+ signals_unblock();
+ const io_wait_ret ret = io_wait(pair, -1, true);
+ signals_block();
+
+ if (ret != IO_WAIT_MORE)
+ goto error;
+ }
+#endif
+
+#ifdef HAVE_POSIX_FADVISE
+ // It will fail with some special files like FIFOs but that is fine.
+ (void)posix_fadvise(pair->src_fd, 0, 0,
+ opt_mode == MODE_LIST
+ ? POSIX_FADV_RANDOM
+ : POSIX_FADV_SEQUENTIAL);
+#endif
+
+ return false;
+
+error_msg:
+ message_error(_("%s: %s"), pair->src_name, strerror(errno));
+error:
+ (void)close(pair->src_fd);
+ return true;
+}
+
+
+extern file_pair *
+io_open_src(const char *src_name)
+{
+ if (src_name[0] == '\0') {
+ message_error(_("Empty filename, skipping"));
+ return NULL;
+ }
+
+ // Since we have only one file open at a time, we can use
+ // a statically allocated structure.
+ static file_pair pair;
+
+ // This implicitly also initializes src_st.st_size to zero
+ // which is expected to be <= 0 by default. fstat() isn't
+ // called when reading from standard input but src_st.st_size
+ // is still read.
+ pair = (file_pair){
+ .src_name = src_name,
+ .dest_name = NULL,
+ .src_fd = -1,
+ .dest_fd = -1,
+ .src_eof = false,
+ .src_has_seen_input = false,
+ .flush_needed = false,
+ .dest_try_sparse = false,
+ .dest_pending_sparse = 0,
+ };
+
+ // Block the signals, for which we have a custom signal handler, so
+ // that we don't need to worry about EINTR.
+ signals_block();
+ const bool error = io_open_src_real(&pair);
+ signals_unblock();
+
+#ifdef ENABLE_SANDBOX
+ if (!error)
+ io_sandbox_enter(pair.src_fd);
+#endif
+
+ return error ? NULL : &pair;
+}
+
+
+/// \brief Closes source file of the file_pair structure
+///
+/// \param pair File whose src_fd should be closed
+/// \param success If true, the file will be removed from the disk if
+/// closing succeeds and --keep hasn't been used.
+static void
+io_close_src(file_pair *pair, bool success)
+{
+#ifndef TUKLIB_DOSLIKE
+ if (restore_stdin_flags) {
+ assert(pair->src_fd == STDIN_FILENO);
+
+ restore_stdin_flags = false;
+
+ if (fcntl(STDIN_FILENO, F_SETFL, stdin_flags) == -1)
+ message_error(_("Error restoring the status flags "
+ "to standard input: %s"),
+ strerror(errno));
+ }
+#endif
+
+ if (pair->src_fd != STDIN_FILENO && pair->src_fd != -1) {
+ // Close the file before possibly unlinking it. On DOS-like
+ // systems this is always required since unlinking will fail
+ // if the file is open. On POSIX systems it usually works
+ // to unlink open files, but in some cases it doesn't and
+ // one gets EBUSY in errno.
+ //
+ // xz 5.2.2 and older unlinked the file before closing it
+ // (except on DOS-like systems). The old code didn't handle
+ // EBUSY and could fail e.g. on some CIFS shares. The
+ // advantage of unlinking before closing is negligible
+ // (avoids a race between close() and stat()/lstat() and
+ // unlink()), so let's keep this simple.
+ (void)close(pair->src_fd);
+
+ if (success && !opt_keep_original)
+ io_unlink(pair->src_name, &pair->src_st);
+ }
+
+ return;
+}
+
+
+static bool
+io_open_dest_real(file_pair *pair)
+{
+ if (opt_stdout || pair->src_fd == STDIN_FILENO) {
+ // We don't modify or free() this.
+ pair->dest_name = (char *)"(stdout)";
+ pair->dest_fd = STDOUT_FILENO;
+#ifdef TUKLIB_DOSLIKE
+ setmode(STDOUT_FILENO, O_BINARY);
+#else
+ // Try to set O_NONBLOCK if it isn't already set.
+ // If it fails, we assume that stdout is non-blocking
+ // in practice. See the comments in io_open_src_real()
+ // for similar situation with stdin.
+ //
+ // NOTE: O_APPEND may be unset later in this function
+ // and it relies on stdout_flags being set here.
+ stdout_flags = fcntl(STDOUT_FILENO, F_GETFL);
+ if (stdout_flags == -1) {
+ message_error(_("Error getting the file status flags "
+ "from standard output: %s"),
+ strerror(errno));
+ return true;
+ }
+
+ if ((stdout_flags & O_NONBLOCK) == 0
+ && fcntl(STDOUT_FILENO, F_SETFL,
+ stdout_flags | O_NONBLOCK) != -1)
+ restore_stdout_flags = true;
+#endif
+ } else {
+ pair->dest_name = suffix_get_dest_name(pair->src_name);
+ if (pair->dest_name == NULL)
+ return true;
+
+#ifdef __DJGPP__
+ struct stat st;
+ if (stat(pair->dest_name, &st) == 0) {
+ // Check that it isn't a special file like "prn".
+ if (st.st_dev == -1) {
+ message_error("%s: Refusing to write to "
+ "a DOS special file",
+ pair->dest_name);
+ free(pair->dest_name);
+ return true;
+ }
+
+ // Check that we aren't overwriting the source file.
+ if (st.st_dev == pair->src_st.st_dev
+ && st.st_ino == pair->src_st.st_ino) {
+ message_error("%s: Output file is the same "
+ "as the input file",
+ pair->dest_name);
+ free(pair->dest_name);
+ return true;
+ }
+ }
+#endif
+
+ // If --force was used, unlink the target file first.
+ if (opt_force && unlink(pair->dest_name) && errno != ENOENT) {
+ message_error(_("%s: Cannot remove: %s"),
+ pair->dest_name, strerror(errno));
+ free(pair->dest_name);
+ return true;
+ }
+
+ // Open the file.
+ int flags = O_WRONLY | O_BINARY | O_NOCTTY
+ | O_CREAT | O_EXCL;
+#ifndef TUKLIB_DOSLIKE
+ flags |= O_NONBLOCK;
+#endif
+ const mode_t mode = S_IRUSR | S_IWUSR;
+ pair->dest_fd = open(pair->dest_name, flags, mode);
+
+ if (pair->dest_fd == -1) {
+ message_error(_("%s: %s"), pair->dest_name,
+ strerror(errno));
+ free(pair->dest_name);
+ return true;
+ }
+ }
+
+ if (fstat(pair->dest_fd, &pair->dest_st)) {
+ // If fstat() really fails, we have a safe fallback here.
+#if defined(__VMS)
+ pair->dest_st.st_ino[0] = 0;
+ pair->dest_st.st_ino[1] = 0;
+ pair->dest_st.st_ino[2] = 0;
+#else
+ pair->dest_st.st_dev = 0;
+ pair->dest_st.st_ino = 0;
+#endif
+ }
+#if defined(TUKLIB_DOSLIKE) && !defined(__DJGPP__)
+ // Check that the output file is a regular file. We open with O_EXCL
+ // but that doesn't prevent open()/_open() on Windows from opening
+ // files like "con" or "nul".
+ //
+ // With DJGPP this check is done with stat() even before opening
+ // the output file. That method or a variant of it doesn't work on
+ // Windows because on Windows stat()/_stat64() sets st.st_mode so
+ // that S_ISREG(st.st_mode) will be true even for special files.
+ // With fstat()/_fstat64() it works.
+ else if (pair->dest_fd != STDOUT_FILENO
+ && !S_ISREG(pair->dest_st.st_mode)) {
+ message_error("%s: Destination is not a regular file",
+ pair->dest_name);
+
+ // dest_fd needs to be reset to -1 to keep io_close() working.
+ (void)close(pair->dest_fd);
+ pair->dest_fd = -1;
+
+ free(pair->dest_name);
+ return true;
+ }
+#elif !defined(TUKLIB_DOSLIKE)
+ else if (try_sparse && opt_mode == MODE_DECOMPRESS) {
+ // When writing to standard output, we need to be extra
+ // careful:
+ // - It may be connected to something else than
+ // a regular file.
+ // - We aren't necessarily writing to a new empty file
+ // or to the end of an existing file.
+ // - O_APPEND may be active.
+ //
+ // TODO: I'm keeping this disabled for DOS-like systems
+ // for now. FAT doesn't support sparse files, but NTFS
+ // does, so maybe this should be enabled on Windows after
+ // some testing.
+ if (pair->dest_fd == STDOUT_FILENO) {
+ if (!S_ISREG(pair->dest_st.st_mode))
+ return false;
+
+ if (stdout_flags & O_APPEND) {
+ // Creating a sparse file is not possible
+ // when O_APPEND is active (it's used by
+ // shell's >> redirection). As I understand
+ // it, it is safe to temporarily disable
+ // O_APPEND in xz, because if someone
+ // happened to write to the same file at the
+ // same time, results would be bad anyway
+ // (users shouldn't assume that xz uses any
+ // specific block size when writing data).
+ //
+ // The write position may be something else
+ // than the end of the file, so we must fix
+ // it to start writing at the end of the file
+ // to imitate O_APPEND.
+ if (lseek(STDOUT_FILENO, 0, SEEK_END) == -1)
+ return false;
+
+ // Construct the new file status flags.
+ // If O_NONBLOCK was set earlier in this
+ // function, it must be kept here too.
+ int flags = stdout_flags & ~O_APPEND;
+ if (restore_stdout_flags)
+ flags |= O_NONBLOCK;
+
+ // If this fcntl() fails, we continue but won't
+ // try to create sparse output. The original
+ // flags will still be restored if needed (to
+ // unset O_NONBLOCK) when the file is finished.
+ if (fcntl(STDOUT_FILENO, F_SETFL, flags) == -1)
+ return false;
+
+ // Disabling O_APPEND succeeded. Mark
+ // that the flags should be restored
+ // in io_close_dest(). (This may have already
+ // been set when enabling O_NONBLOCK.)
+ restore_stdout_flags = true;
+
+ } else if (lseek(STDOUT_FILENO, 0, SEEK_CUR)
+ != pair->dest_st.st_size) {
+ // Writing won't start exactly at the end
+ // of the file. We cannot use sparse output,
+ // because it would probably corrupt the file.
+ return false;
+ }
+ }
+
+ pair->dest_try_sparse = true;
+ }
+#endif
+
+ return false;
+}
+
+
+extern bool
+io_open_dest(file_pair *pair)
+{
+ signals_block();
+ const bool ret = io_open_dest_real(pair);
+ signals_unblock();
+ return ret;
+}
+
+
+/// \brief Closes destination file of the file_pair structure
+///
+/// \param pair File whose dest_fd should be closed
+/// \param success If false, the file will be removed from the disk.
+///
+/// \return Zero if closing succeeds. On error, -1 is returned and
+/// error message printed.
+static bool
+io_close_dest(file_pair *pair, bool success)
+{
+#ifndef TUKLIB_DOSLIKE
+ // If io_open_dest() has disabled O_APPEND, restore it here.
+ if (restore_stdout_flags) {
+ assert(pair->dest_fd == STDOUT_FILENO);
+
+ restore_stdout_flags = false;
+
+ if (fcntl(STDOUT_FILENO, F_SETFL, stdout_flags) == -1) {
+ message_error(_("Error restoring the O_APPEND flag "
+ "to standard output: %s"),
+ strerror(errno));
+ return true;
+ }
+ }
+#endif
+
+ if (pair->dest_fd == -1 || pair->dest_fd == STDOUT_FILENO)
+ return false;
+
+ if (close(pair->dest_fd)) {
+ message_error(_("%s: Closing the file failed: %s"),
+ pair->dest_name, strerror(errno));
+
+ // Closing destination file failed, so we cannot trust its
+ // contents. Get rid of junk:
+ io_unlink(pair->dest_name, &pair->dest_st);
+ free(pair->dest_name);
+ return true;
+ }
+
+ // If the operation using this file wasn't successful, we git rid
+ // of the junk file.
+ if (!success)
+ io_unlink(pair->dest_name, &pair->dest_st);
+
+ free(pair->dest_name);
+
+ return false;
+}
+
+
+extern void
+io_close(file_pair *pair, bool success)
+{
+ // Take care of sparseness at the end of the output file.
+ if (success && pair->dest_try_sparse
+ && pair->dest_pending_sparse > 0) {
+ // Seek forward one byte less than the size of the pending
+ // hole, then write one zero-byte. This way the file grows
+ // to its correct size. An alternative would be to use
+ // ftruncate() but that isn't portable enough (e.g. it
+ // doesn't work with FAT on Linux; FAT isn't that important
+ // since it doesn't support sparse files anyway, but we don't
+ // want to create corrupt files on it).
+ if (lseek(pair->dest_fd, pair->dest_pending_sparse - 1,
+ SEEK_CUR) == -1) {
+ message_error(_("%s: Seeking failed when trying "
+ "to create a sparse file: %s"),
+ pair->dest_name, strerror(errno));
+ success = false;
+ } else {
+ const uint8_t zero[1] = { '\0' };
+ if (io_write_buf(pair, zero, 1))
+ success = false;
+ }
+ }
+
+ signals_block();
+
+ // Copy the file attributes. We need to skip this if destination
+ // file isn't open or it is standard output.
+ if (success && pair->dest_fd != -1 && pair->dest_fd != STDOUT_FILENO)
+ io_copy_attrs(pair);
+
+ // Close the destination first. If it fails, we must not remove
+ // the source file!
+ if (io_close_dest(pair, success))
+ success = false;
+
+ // Close the source file, and unlink it if the operation using this
+ // file pair was successful and we haven't requested to keep the
+ // source file.
+ io_close_src(pair, success);
+
+ signals_unblock();
+
+ return;
+}
+
+
+extern void
+io_fix_src_pos(file_pair *pair, size_t rewind_size)
+{
+ assert(rewind_size <= IO_BUFFER_SIZE);
+
+ if (rewind_size > 0) {
+ // This doesn't need to work on unseekable file descriptors,
+ // so just ignore possible errors.
+ (void)lseek(pair->src_fd, -(off_t)(rewind_size), SEEK_CUR);
+ }
+
+ return;
+}
+
+
+extern size_t
+io_read(file_pair *pair, io_buf *buf, size_t size)
+{
+ assert(size <= IO_BUFFER_SIZE);
+
+ size_t pos = 0;
+
+ while (pos < size) {
+ const ssize_t amount = read(
+ pair->src_fd, buf->u8 + pos, size - pos);
+
+ if (amount == 0) {
+ pair->src_eof = true;
+ break;
+ }
+
+ if (amount == -1) {
+ if (errno == EINTR) {
+ if (user_abort)
+ return SIZE_MAX;
+
+ continue;
+ }
+
+#ifndef TUKLIB_DOSLIKE
+ if (IS_EAGAIN_OR_EWOULDBLOCK(errno)) {
+ // Disable the flush-timeout if no input has
+ // been seen since the previous flush and thus
+ // there would be nothing to flush after the
+ // timeout expires (avoids busy waiting).
+ const int timeout = pair->src_has_seen_input
+ ? mytime_get_flush_timeout()
+ : -1;
+
+ switch (io_wait(pair, timeout, true)) {
+ case IO_WAIT_MORE:
+ continue;
+
+ case IO_WAIT_ERROR:
+ return SIZE_MAX;
+
+ case IO_WAIT_TIMEOUT:
+ pair->flush_needed = true;
+ return pos;
+
+ default:
+ message_bug();
+ }
+ }
+#endif
+
+ message_error(_("%s: Read error: %s"),
+ pair->src_name, strerror(errno));
+
+ return SIZE_MAX;
+ }
+
+ pos += (size_t)(amount);
+
+ if (!pair->src_has_seen_input) {
+ pair->src_has_seen_input = true;
+ mytime_set_flush_time();
+ }
+ }
+
+ return pos;
+}
+
+
+extern bool
+io_seek_src(file_pair *pair, uint64_t pos)
+{
+ // Caller must not attempt to seek past the end of the input file
+ // (seeking to 100 in a 100-byte file is seeking to the end of
+ // the file, not past the end of the file, and thus that is allowed).
+ //
+ // This also validates that pos can be safely cast to off_t.
+ if (pos > (uint64_t)(pair->src_st.st_size))
+ message_bug();
+
+ if (lseek(pair->src_fd, (off_t)(pos), SEEK_SET) == -1) {
+ message_error(_("%s: Error seeking the file: %s"),
+ pair->src_name, strerror(errno));
+ return true;
+ }
+
+ pair->src_eof = false;
+
+ return false;
+}
+
+
+extern bool
+io_pread(file_pair *pair, io_buf *buf, size_t size, uint64_t pos)
+{
+ // Using lseek() and read() is more portable than pread() and
+ // for us it is as good as real pread().
+ if (io_seek_src(pair, pos))
+ return true;
+
+ const size_t amount = io_read(pair, buf, size);
+ if (amount == SIZE_MAX)
+ return true;
+
+ if (amount != size) {
+ message_error(_("%s: Unexpected end of file"),
+ pair->src_name);
+ return true;
+ }
+
+ return false;
+}
+
+
+static bool
+is_sparse(const io_buf *buf)
+{
+ assert(IO_BUFFER_SIZE % sizeof(uint64_t) == 0);
+
+ for (size_t i = 0; i < ARRAY_SIZE(buf->u64); ++i)
+ if (buf->u64[i] != 0)
+ return false;
+
+ return true;
+}
+
+
+static bool
+io_write_buf(file_pair *pair, const uint8_t *buf, size_t size)
+{
+ assert(size <= IO_BUFFER_SIZE);
+
+ while (size > 0) {
+ const ssize_t amount = write(pair->dest_fd, buf, size);
+ if (amount == -1) {
+ if (errno == EINTR) {
+ if (user_abort)
+ return true;
+
+ continue;
+ }
+
+#ifndef TUKLIB_DOSLIKE
+ if (IS_EAGAIN_OR_EWOULDBLOCK(errno)) {
+ if (io_wait(pair, -1, false) == IO_WAIT_MORE)
+ continue;
+
+ return true;
+ }
+#endif
+
+ // Handle broken pipe specially. gzip and bzip2
+ // don't print anything on SIGPIPE. In addition,
+ // gzip --quiet uses exit status 2 (warning) on
+ // broken pipe instead of whatever raise(SIGPIPE)
+ // would make it return. It is there to hide "Broken
+ // pipe" message on some old shells (probably old
+ // GNU bash).
+ //
+ // We don't do anything special with --quiet, which
+ // is what bzip2 does too. If we get SIGPIPE, we
+ // will handle it like other signals by setting
+ // user_abort, and get EPIPE here.
+ if (errno != EPIPE)
+ message_error(_("%s: Write error: %s"),
+ pair->dest_name, strerror(errno));
+
+ return true;
+ }
+
+ buf += (size_t)(amount);
+ size -= (size_t)(amount);
+ }
+
+ return false;
+}
+
+
+extern bool
+io_write(file_pair *pair, const io_buf *buf, size_t size)
+{
+ assert(size <= IO_BUFFER_SIZE);
+
+ if (pair->dest_try_sparse) {
+ // Check if the block is sparse (contains only zeros). If it
+ // sparse, we just store the amount and return. We will take
+ // care of actually skipping over the hole when we hit the
+ // next data block or close the file.
+ //
+ // Since io_close() requires that dest_pending_sparse > 0
+ // if the file ends with sparse block, we must also return
+ // if size == 0 to avoid doing the lseek().
+ if (size == IO_BUFFER_SIZE) {
+ // Even if the block was sparse, treat it as non-sparse
+ // if the pending sparse amount is large compared to
+ // the size of off_t. In practice this only matters
+ // on 32-bit systems where off_t isn't always 64 bits.
+ const off_t pending_max
+ = (off_t)(1) << (sizeof(off_t) * CHAR_BIT - 2);
+ if (is_sparse(buf) && pair->dest_pending_sparse
+ < pending_max) {
+ pair->dest_pending_sparse += (off_t)(size);
+ return false;
+ }
+ } else if (size == 0) {
+ return false;
+ }
+
+ // This is not a sparse block. If we have a pending hole,
+ // skip it now.
+ if (pair->dest_pending_sparse > 0) {
+ if (lseek(pair->dest_fd, pair->dest_pending_sparse,
+ SEEK_CUR) == -1) {
+ message_error(_("%s: Seeking failed when "
+ "trying to create a sparse "
+ "file: %s"), pair->dest_name,
+ strerror(errno));
+ return true;
+ }
+
+ pair->dest_pending_sparse = 0;
+ }
+ }
+
+ return io_write_buf(pair, buf->u8, size);
+}
diff --git a/src/xz/file_io.h b/src/xz/file_io.h
new file mode 100644
index 0000000..6992efa
--- /dev/null
+++ b/src/xz/file_io.h
@@ -0,0 +1,179 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file file_io.h
+/// \brief I/O types and functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+// Some systems have suboptimal BUFSIZ. Use a bit bigger value on them.
+// We also need that IO_BUFFER_SIZE is a multiple of 8 (sizeof(uint64_t))
+#if BUFSIZ <= 1024
+# define IO_BUFFER_SIZE 8192
+#else
+# define IO_BUFFER_SIZE (BUFSIZ & ~7U)
+#endif
+
+
+/// is_sparse() accesses the buffer as uint64_t for maximum speed.
+/// The u32 and u64 members must only be access through this union
+/// to avoid strict aliasing violations. Taking a pointer of u8
+/// should be fine as long as uint8_t maps to unsigned char which
+/// can alias anything.
+typedef union {
+ uint8_t u8[IO_BUFFER_SIZE];
+ uint32_t u32[IO_BUFFER_SIZE / sizeof(uint32_t)];
+ uint64_t u64[IO_BUFFER_SIZE / sizeof(uint64_t)];
+} io_buf;
+
+
+typedef struct {
+ /// Name of the source filename (as given on the command line) or
+ /// pointer to static "(stdin)" when reading from standard input.
+ const char *src_name;
+
+ /// Destination filename converted from src_name or pointer to static
+ /// "(stdout)" when writing to standard output.
+ char *dest_name;
+
+ /// File descriptor of the source file
+ int src_fd;
+
+ /// File descriptor of the target file
+ int dest_fd;
+
+ /// True once end of the source file has been detected.
+ bool src_eof;
+
+ /// For --flush-timeout: True if at least one byte has been read
+ /// since the previous flush or the start of the file.
+ bool src_has_seen_input;
+
+ /// For --flush-timeout: True when flushing is needed.
+ bool flush_needed;
+
+ /// If true, we look for long chunks of zeros and try to create
+ /// a sparse file.
+ bool dest_try_sparse;
+
+ /// This is used only if dest_try_sparse is true. This holds the
+ /// number of zero bytes we haven't written out, because we plan
+ /// to make that byte range a sparse chunk.
+ off_t dest_pending_sparse;
+
+ /// Stat of the source file.
+ struct stat src_st;
+
+ /// Stat of the destination file.
+ struct stat dest_st;
+
+} file_pair;
+
+
+/// \brief Initialize the I/O module
+extern void io_init(void);
+
+
+#ifndef TUKLIB_DOSLIKE
+/// \brief Write a byte to user_abort_pipe[1]
+///
+/// This is called from a signal handler.
+extern void io_write_to_user_abort_pipe(void);
+#endif
+
+
+/// \brief Disable creation of sparse files when decompressing
+extern void io_no_sparse(void);
+
+
+#ifdef ENABLE_SANDBOX
+/// \brief main() calls this if conditions for sandboxing have been met.
+extern void io_allow_sandbox(void);
+#endif
+
+
+/// \brief Open the source file
+extern file_pair *io_open_src(const char *src_name);
+
+
+/// \brief Open the destination file
+extern bool io_open_dest(file_pair *pair);
+
+
+/// \brief Closes the file descriptors and frees possible allocated memory
+///
+/// The success argument determines if source or destination file gets
+/// unlinked:
+/// - false: The destination file is unlinked.
+/// - true: The source file is unlinked unless writing to stdout or --keep
+/// was used.
+extern void io_close(file_pair *pair, bool success);
+
+
+/// \brief Reads from the source file to a buffer
+///
+/// \param pair File pair having the source file open for reading
+/// \param buf Destination buffer to hold the read data
+/// \param size Size of the buffer; must be at most IO_BUFFER_SIZE
+///
+/// \return On success, number of bytes read is returned. On end of
+/// file zero is returned and pair->src_eof set to true.
+/// On error, SIZE_MAX is returned and error message printed.
+extern size_t io_read(file_pair *pair, io_buf *buf, size_t size);
+
+
+/// \brief Fix the position in src_fd
+///
+/// This is used when --single-thream has been specified and decompression
+/// is successful. If the input file descriptor supports seeking, this
+/// function fixes the input position to point to the next byte after the
+/// decompressed stream.
+///
+/// \param pair File pair having the source file open for reading
+/// \param rewind_size How many bytes of extra have been read i.e.
+/// how much to seek backwards.
+extern void io_fix_src_pos(file_pair *pair, size_t rewind_size);
+
+
+/// \brief Seek to the given absolute position in the source file
+///
+/// This calls lseek() and also clears pair->src_eof.
+///
+/// \param pair Seekable source file
+/// \param pos Offset relative to the beginning of the file,
+/// from which the data should be read.
+///
+/// \return On success, false is returned. On error, error message
+/// is printed and true is returned.
+extern bool io_seek_src(file_pair *pair, uint64_t pos);
+
+
+/// \brief Read from source file from given offset to a buffer
+///
+/// This is remotely similar to standard pread(). This uses lseek() though,
+/// so the read offset is changed on each call.
+///
+/// \param pair Seekable source file
+/// \param buf Destination buffer
+/// \param size Amount of data to read
+/// \param pos Offset relative to the beginning of the file,
+/// from which the data should be read.
+///
+/// \return On success, false is returned. On error, error message
+/// is printed and true is returned.
+extern bool io_pread(file_pair *pair, io_buf *buf, size_t size, uint64_t pos);
+
+
+/// \brief Writes a buffer to the destination file
+///
+/// \param pair File pair having the destination file open for writing
+/// \param buf Buffer containing the data to be written
+/// \param size Size of the buffer; must be at most IO_BUFFER_SIZE
+///
+/// \return On success, zero is returned. On error, -1 is returned
+/// and error message printed.
+extern bool io_write(file_pair *pair, const io_buf *buf, size_t size);
diff --git a/src/xz/hardware.c b/src/xz/hardware.c
new file mode 100644
index 0000000..c694882
--- /dev/null
+++ b/src/xz/hardware.c
@@ -0,0 +1,338 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file hardware.c
+/// \brief Detection of available hardware resources
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+
+/// Maximum number of worker threads. This can be set with
+/// the --threads=NUM command line option.
+static uint32_t threads_max = 1;
+
+/// True when the number of threads is automatically determined based
+/// on the available hardware threads.
+static bool threads_are_automatic = false;
+
+/// If true, then try to use multi-threaded mode (if memlimit allows)
+/// even if only one thread was requested explicitly (-T+1).
+static bool use_mt_mode_with_one_thread = false;
+
+/// Memory usage limit for compression
+static uint64_t memlimit_compress = 0;
+
+/// Memory usage limit for decompression
+static uint64_t memlimit_decompress = 0;
+
+/// Default memory usage for multithreaded modes:
+///
+/// - Default value for --memlimit-compress when automatic number of threads
+/// is used. However, if the limit wouldn't allow even one thread then
+/// the limit is ignored in coder.c and one thread will be used anyway.
+/// This mess is a compromise: we wish to prevent -T0 from using too
+/// many threads but we also don't want xz to give an error due to
+/// a memlimit that the user didn't explicitly set.
+///
+/// - Default value for --memlimit-mt-decompress
+///
+/// This value is calculated in hardware_init() and cannot be changed later.
+static uint64_t memlimit_mt_default;
+
+/// Memory usage limit for multithreaded decompression. This is a soft limit:
+/// if reducing the number of threads to one isn't enough to keep memory
+/// usage below this limit, then one thread is used and this limit is ignored.
+/// memlimit_decompress is still obeyed.
+///
+/// This can be set with --memlimit-mt-decompress. The default value for
+/// this is memlimit_mt_default.
+static uint64_t memlimit_mtdec;
+
+/// Total amount of physical RAM
+static uint64_t total_ram;
+
+
+extern void
+hardware_threads_set(uint32_t n)
+{
+ // Reset these to false first and set them to true when appropriate.
+ threads_are_automatic = false;
+ use_mt_mode_with_one_thread = false;
+
+ if (n == 0) {
+ // Automatic number of threads was requested.
+ // If there is only one hardware thread, multi-threaded
+ // mode will still be used if memory limit allows.
+ threads_are_automatic = true;
+ use_mt_mode_with_one_thread = true;
+
+ // If threading support was enabled at build time,
+ // use the number of available CPU cores. Otherwise
+ // use one thread since disabling threading support
+ // omits lzma_cputhreads() from liblzma.
+#ifdef MYTHREAD_ENABLED
+ threads_max = lzma_cputhreads();
+ if (threads_max == 0)
+ threads_max = 1;
+#else
+ threads_max = 1;
+#endif
+ } else if (n == UINT32_MAX) {
+ use_mt_mode_with_one_thread = true;
+ threads_max = 1;
+ } else {
+ threads_max = n;
+ }
+
+ return;
+}
+
+
+extern uint32_t
+hardware_threads_get(void)
+{
+ return threads_max;
+}
+
+
+extern bool
+hardware_threads_is_mt(void)
+{
+#ifdef MYTHREAD_ENABLED
+ return threads_max > 1 || use_mt_mode_with_one_thread;
+#else
+ return false;
+#endif
+}
+
+
+extern void
+hardware_memlimit_set(uint64_t new_memlimit,
+ bool set_compress, bool set_decompress, bool set_mtdec,
+ bool is_percentage)
+{
+ if (is_percentage) {
+ assert(new_memlimit > 0);
+ assert(new_memlimit <= 100);
+ new_memlimit = (uint32_t)new_memlimit * total_ram / 100;
+ }
+
+ if (set_compress) {
+ memlimit_compress = new_memlimit;
+
+#if SIZE_MAX == UINT32_MAX
+ // FIXME?
+ //
+ // When running a 32-bit xz on a system with a lot of RAM and
+ // using a percentage-based memory limit, the result can be
+ // bigger than the 32-bit address space. Limiting the limit
+ // below SIZE_MAX for compression (not decompression) makes
+ // xz lower the compression settings (or number of threads)
+ // to a level that *might* work. In practice it has worked
+ // when using a 64-bit kernel that gives full 4 GiB address
+ // space to 32-bit programs. In other situations this might
+ // still be too high, like 32-bit kernels that may give much
+ // less than 4 GiB to a single application.
+ //
+ // So this is an ugly hack but I will keep it here while
+ // it does more good than bad.
+ //
+ // Use a value less than SIZE_MAX so that there's some room
+ // for the xz program and so on. Don't use 4000 MiB because
+ // it could look like someone mixed up base-2 and base-10.
+#ifdef __mips__
+ // For MIPS32, due to architectural peculiarities,
+ // the limit is even lower.
+ const uint64_t limit_max = UINT64_C(2000) << 20;
+#else
+ const uint64_t limit_max = UINT64_C(4020) << 20;
+#endif
+
+ // UINT64_MAX is a special case for the string "max" so
+ // that has to be handled specially.
+ if (memlimit_compress != UINT64_MAX
+ && memlimit_compress > limit_max)
+ memlimit_compress = limit_max;
+#endif
+ }
+
+ if (set_decompress)
+ memlimit_decompress = new_memlimit;
+
+ if (set_mtdec)
+ memlimit_mtdec = new_memlimit;
+
+ return;
+}
+
+
+extern uint64_t
+hardware_memlimit_get(enum operation_mode mode)
+{
+ // 0 is a special value that indicates the default.
+ // It disables the limit in single-threaded mode.
+ //
+ // NOTE: For multithreaded decompression, this is the hard limit
+ // (memlimit_stop). hardware_memlimit_mtdec_get() gives the
+ // soft limit (memlimit_threaded).
+ const uint64_t memlimit = mode == MODE_COMPRESS
+ ? memlimit_compress : memlimit_decompress;
+ return memlimit != 0 ? memlimit : UINT64_MAX;
+}
+
+
+extern uint64_t
+hardware_memlimit_mtenc_get(void)
+{
+ return hardware_memlimit_mtenc_is_default()
+ ? memlimit_mt_default
+ : hardware_memlimit_get(MODE_COMPRESS);
+}
+
+
+extern bool
+hardware_memlimit_mtenc_is_default(void)
+{
+ return memlimit_compress == 0 && threads_are_automatic;
+}
+
+
+extern uint64_t
+hardware_memlimit_mtdec_get(void)
+{
+ uint64_t m = memlimit_mtdec != 0
+ ? memlimit_mtdec
+ : memlimit_mt_default;
+
+ // Cap the value to memlimit_decompress if it has been specified.
+ // This is nice for --info-memory. It wouldn't be needed for liblzma
+ // since it does this anyway.
+ if (memlimit_decompress != 0 && m > memlimit_decompress)
+ m = memlimit_decompress;
+
+ return m;
+}
+
+
+/// Helper for hardware_memlimit_show() to print one human-readable info line.
+static void
+memlimit_show(const char *str, size_t str_columns, uint64_t value)
+{
+ // Calculate the field width so that str will be padded to take
+ // str_columns on the terminal.
+ //
+ // NOTE: If the string is invalid, this will be -1. Using -1 as
+ // the field width is fine here so it's not handled specially.
+ const int fw = tuklib_mbstr_fw(str, (int)(str_columns));
+
+ // The memory usage limit is considered to be disabled if value
+ // is 0 or UINT64_MAX. This might get a bit more complex once there
+ // is threading support. See the comment in hardware_memlimit_get().
+ if (value == 0 || value == UINT64_MAX)
+ printf(" %-*s %s\n", fw, str, _("Disabled"));
+ else
+ printf(" %-*s %s MiB (%s B)\n", fw, str,
+ uint64_to_str(round_up_to_mib(value), 0),
+ uint64_to_str(value, 1));
+
+ return;
+}
+
+
+extern void
+hardware_memlimit_show(void)
+{
+ uint32_t cputhreads = 1;
+#ifdef MYTHREAD_ENABLED
+ cputhreads = lzma_cputhreads();
+ if (cputhreads == 0)
+ cputhreads = 1;
+#endif
+
+ if (opt_robot) {
+ printf("%" PRIu64 "\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%" PRIu64 "\t%" PRIu32 "\n",
+ total_ram,
+ memlimit_compress,
+ memlimit_decompress,
+ hardware_memlimit_mtdec_get(),
+ memlimit_mt_default,
+ cputhreads);
+ } else {
+ const char *msgs[] = {
+ _("Amount of physical memory (RAM):"),
+ _("Number of processor threads:"),
+ _("Compression:"),
+ _("Decompression:"),
+ _("Multi-threaded decompression:"),
+ _("Default for -T0:"),
+ };
+
+ size_t width_max = 1;
+ for (unsigned i = 0; i < ARRAY_SIZE(msgs); ++i) {
+ size_t w = tuklib_mbstr_width(msgs[i], NULL);
+
+ // When debugging, catch invalid strings with
+ // an assertion. Otherwise fallback to 1 so
+ // that the columns just won't be aligned.
+ assert(w != (size_t)-1);
+ if (w == (size_t)-1)
+ w = 1;
+
+ if (width_max < w)
+ width_max = w;
+ }
+
+ puts(_("Hardware information:"));
+ memlimit_show(msgs[0], width_max, total_ram);
+ printf(" %-*s %" PRIu32 "\n",
+ tuklib_mbstr_fw(msgs[1], (int)(width_max)),
+ msgs[1], cputhreads);
+
+ putchar('\n');
+ puts(_("Memory usage limits:"));
+ memlimit_show(msgs[2], width_max, memlimit_compress);
+ memlimit_show(msgs[3], width_max, memlimit_decompress);
+ memlimit_show(msgs[4], width_max,
+ hardware_memlimit_mtdec_get());
+ memlimit_show(msgs[5], width_max, memlimit_mt_default);
+ }
+
+ tuklib_exit(E_SUCCESS, E_ERROR, message_verbosity_get() != V_SILENT);
+}
+
+
+extern void
+hardware_init(void)
+{
+ // Get the amount of RAM. If we cannot determine it,
+ // use the assumption defined by the configure script.
+ total_ram = lzma_physmem();
+ if (total_ram == 0)
+ total_ram = (uint64_t)(ASSUME_RAM) * 1024 * 1024;
+
+ // FIXME? There may be better methods to determine the default value.
+ // One Linux-specific suggestion is to use MemAvailable from
+ // /proc/meminfo as the starting point.
+ memlimit_mt_default = total_ram / 4;
+
+#if SIZE_MAX == UINT32_MAX
+ // A too high value may cause 32-bit xz to run out of address space.
+ // Use a conservative maximum value here. A few typical address space
+ // sizes with Linux:
+ // - x86-64 with 32-bit xz: 4 GiB
+ // - x86: 3 GiB
+ // - MIPS32: 2 GiB
+ const size_t mem_ceiling = 1400U << 20;
+ if (memlimit_mt_default > mem_ceiling)
+ memlimit_mt_default = mem_ceiling;
+#endif
+
+ return;
+}
diff --git a/src/xz/hardware.h b/src/xz/hardware.h
new file mode 100644
index 0000000..a67b26e
--- /dev/null
+++ b/src/xz/hardware.h
@@ -0,0 +1,75 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file hardware.h
+/// \brief Detection of available hardware resources
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// Initialize some hardware-specific variables, which are needed by other
+/// hardware_* functions.
+extern void hardware_init(void);
+
+
+/// Set the maximum number of worker threads.
+/// A special value of UINT32_MAX sets one thread in multi-threaded mode.
+extern void hardware_threads_set(uint32_t threadlimit);
+
+/// Get the maximum number of worker threads.
+extern uint32_t hardware_threads_get(void);
+
+/// Returns true if multithreaded mode should be used for .xz compression.
+/// This can be true even if the number of threads is one.
+extern bool hardware_threads_is_mt(void);
+
+
+/// Set the memory usage limit. There are separate limits for compression,
+/// decompression (also includes --list), and multithreaded decompression.
+/// Any combination of these can be set with a single call to this function.
+/// Zero indicates resetting the limit back to the defaults.
+/// The limit can also be set as a percentage of installed RAM; the
+/// percentage must be in the range [1, 100].
+extern void hardware_memlimit_set(uint64_t new_memlimit,
+ bool set_compress, bool set_decompress, bool set_mtdec,
+ bool is_percentage);
+
+/// Get the current memory usage limit for compression or decompression.
+/// This is a hard limit that will not be exceeded. This is obeyed in
+/// both single-threaded and multithreaded modes.
+extern uint64_t hardware_memlimit_get(enum operation_mode mode);
+
+/// This returns a system-specific default value if all of the following
+/// conditions are true:
+///
+/// - An automatic number of threads was requested (--threads=0).
+///
+/// - --memlimit-compress wasn't used or it was reset to the default
+/// value by setting it to 0.
+///
+/// Otherwise this is identical to hardware_memlimit_get(MODE_COMPRESS).
+///
+/// The idea is to keep automatic thread count reasonable so that too
+/// high memory usage is avoided and, with 32-bit xz, running out of
+/// address space is avoided.
+extern uint64_t hardware_memlimit_mtenc_get(void);
+
+/// Returns true if the value returned by hardware_memlimit_mtenc_get() is
+/// a system-specific default value. coder.c uses this to ignore the default
+/// memlimit in case it's too small even for a single thread in multithreaded
+/// mode. This way the default limit will never make xz fail or affect the
+/// compressed output; it will only make xz reduce the number of threads.
+extern bool hardware_memlimit_mtenc_is_default(void);
+
+/// Get the current memory usage limit for multithreaded decompression.
+/// This is only used to reduce the number of threads. This limit can be
+/// exceeded if the number of threads are reduce to one. Then the value
+/// from hardware_memlimit_get() will be honored like in single-threaded mode.
+extern uint64_t hardware_memlimit_mtdec_get(void);
+
+/// Display the amount of RAM and memory usage limits and exit.
+tuklib_attr_noreturn
+extern void hardware_memlimit_show(void);
diff --git a/src/xz/list.c b/src/xz/list.c
new file mode 100644
index 0000000..86c3a76
--- /dev/null
+++ b/src/xz/list.c
@@ -0,0 +1,1318 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file list.c
+/// \brief Listing information about .xz files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+#include "tuklib_integer.h"
+
+
+/// Information about a .xz file
+typedef struct {
+ /// Combined Index of all Streams in the file
+ lzma_index *idx;
+
+ /// Total amount of Stream Padding
+ uint64_t stream_padding;
+
+ /// Highest memory usage so far
+ uint64_t memusage_max;
+
+ /// True if all Blocks so far have Compressed Size and
+ /// Uncompressed Size fields
+ bool all_have_sizes;
+
+ /// Oldest XZ Utils version that will decompress the file
+ uint32_t min_version;
+
+} xz_file_info;
+
+#define XZ_FILE_INFO_INIT { NULL, 0, 0, true, 50000002 }
+
+
+/// Information about a .xz Block
+typedef struct {
+ /// Size of the Block Header
+ uint32_t header_size;
+
+ /// A few of the Block Flags as a string
+ char flags[3];
+
+ /// Size of the Compressed Data field in the Block
+ lzma_vli compressed_size;
+
+ /// Decoder memory usage for this Block
+ uint64_t memusage;
+
+ /// The filter chain of this Block in human-readable form
+ char *filter_chain;
+
+} block_header_info;
+
+#define BLOCK_HEADER_INFO_INIT { .filter_chain = NULL }
+#define block_header_info_end(bhi) free((bhi)->filter_chain)
+
+
+/// Strings ending in a colon. These are used for lines like
+/// " Foo: 123 MiB". These are grouped because translated strings
+/// may have different maximum string length, and we want to pad all
+/// strings so that the values are aligned nicely.
+static const char *colon_strs[] = {
+ N_("Streams:"),
+ N_("Blocks:"),
+ N_("Compressed size:"),
+ N_("Uncompressed size:"),
+ N_("Ratio:"),
+ N_("Check:"),
+ N_("Stream Padding:"),
+ N_("Memory needed:"),
+ N_("Sizes in headers:"),
+ // This won't be aligned because it's so long:
+ //N_("Minimum XZ Utils version:"),
+ N_("Number of files:"),
+};
+
+/// Enum matching the above strings.
+enum {
+ COLON_STR_STREAMS,
+ COLON_STR_BLOCKS,
+ COLON_STR_COMPRESSED_SIZE,
+ COLON_STR_UNCOMPRESSED_SIZE,
+ COLON_STR_RATIO,
+ COLON_STR_CHECK,
+ COLON_STR_STREAM_PADDING,
+ COLON_STR_MEMORY_NEEDED,
+ COLON_STR_SIZES_IN_HEADERS,
+ //COLON_STR_MINIMUM_XZ_VERSION,
+ COLON_STR_NUMBER_OF_FILES,
+};
+
+/// Field widths to use with printf to pad the strings to use the same number
+/// of columns on a terminal.
+static int colon_strs_fw[ARRAY_SIZE(colon_strs)];
+
+/// Convenience macro to get the translated string and its field width
+/// using a COLON_STR_foo enum.
+#define COLON_STR(num) colon_strs_fw[num], _(colon_strs[num])
+
+
+/// Column headings
+static struct {
+ /// Table column heading string
+ const char *str;
+
+ /// Number of terminal-columns to use for this table-column.
+ /// If a translated string is longer than the initial value,
+ /// this value will be increased in init_headings().
+ int columns;
+
+ /// Field width to use for printf() to pad "str" to use "columns"
+ /// number of columns on a terminal. This is calculated in
+ /// init_headings().
+ int fw;
+
+} headings[] = {
+ { N_("Stream"), 6, 0 },
+ { N_("Block"), 9, 0 },
+ { N_("Blocks"), 9, 0 },
+ { N_("CompOffset"), 15, 0 },
+ { N_("UncompOffset"), 15, 0 },
+ { N_("CompSize"), 15, 0 },
+ { N_("UncompSize"), 15, 0 },
+ { N_("TotalSize"), 15, 0 },
+ { N_("Ratio"), 5, 0 },
+ { N_("Check"), 10, 0 },
+ { N_("CheckVal"), 1, 0 },
+ { N_("Padding"), 7, 0 },
+ { N_("Header"), 5, 0 },
+ { N_("Flags"), 2, 0 },
+ { N_("MemUsage"), 7 + 4, 0 }, // +4 is for " MiB"
+ { N_("Filters"), 1, 0 },
+};
+
+/// Enum matching the above strings.
+enum {
+ HEADING_STREAM,
+ HEADING_BLOCK,
+ HEADING_BLOCKS,
+ HEADING_COMPOFFSET,
+ HEADING_UNCOMPOFFSET,
+ HEADING_COMPSIZE,
+ HEADING_UNCOMPSIZE,
+ HEADING_TOTALSIZE,
+ HEADING_RATIO,
+ HEADING_CHECK,
+ HEADING_CHECKVAL,
+ HEADING_PADDING,
+ HEADING_HEADERSIZE,
+ HEADING_HEADERFLAGS,
+ HEADING_MEMUSAGE,
+ HEADING_FILTERS,
+};
+
+#define HEADING_STR(num) headings[num].fw, _(headings[num].str)
+
+
+/// Check ID to string mapping
+static const char check_names[LZMA_CHECK_ID_MAX + 1][12] = {
+ // TRANSLATORS: Indicates that there is no integrity check.
+ // This string is used in tables. In older xz version this
+ // string was limited to ten columns in a fixed-width font, but
+ // nowadays there is no strict length restriction anymore.
+ N_("None"),
+ "CRC32",
+ // TRANSLATORS: Indicates that integrity check name is not known,
+ // but the Check ID is known (here 2). In older xz version these
+ // strings were limited to ten columns in a fixed-width font, but
+ // nowadays there is no strict length restriction anymore.
+ N_("Unknown-2"),
+ N_("Unknown-3"),
+ "CRC64",
+ N_("Unknown-5"),
+ N_("Unknown-6"),
+ N_("Unknown-7"),
+ N_("Unknown-8"),
+ N_("Unknown-9"),
+ "SHA-256",
+ N_("Unknown-11"),
+ N_("Unknown-12"),
+ N_("Unknown-13"),
+ N_("Unknown-14"),
+ N_("Unknown-15"),
+};
+
+/// Buffer size for get_check_names(). This may be a bit ridiculous,
+/// but at least it's enough if some language needs many multibyte chars.
+#define CHECKS_STR_SIZE 1024
+
+
+/// Value of the Check field as hexadecimal string.
+/// This is set by parse_check_value().
+static char check_value[2 * LZMA_CHECK_SIZE_MAX + 1];
+
+
+/// Totals that are displayed if there was more than one file.
+/// The "files" counter is also used in print_info_adv() to show
+/// the file number.
+static struct {
+ uint64_t files;
+ uint64_t streams;
+ uint64_t blocks;
+ uint64_t compressed_size;
+ uint64_t uncompressed_size;
+ uint64_t stream_padding;
+ uint64_t memusage_max;
+ uint32_t checks;
+ uint32_t min_version;
+ bool all_have_sizes;
+} totals = { 0, 0, 0, 0, 0, 0, 0, 0, 50000002, true };
+
+
+/// Initialize colon_strs_fw[].
+static void
+init_colon_strs(void)
+{
+ // Lengths of translated strings as bytes.
+ size_t lens[ARRAY_SIZE(colon_strs)];
+
+ // Lengths of translated strings as columns.
+ size_t widths[ARRAY_SIZE(colon_strs)];
+
+ // Maximum number of columns needed by a translated string.
+ size_t width_max = 0;
+
+ for (unsigned i = 0; i < ARRAY_SIZE(colon_strs); ++i) {
+ widths[i] = tuklib_mbstr_width(_(colon_strs[i]), &lens[i]);
+
+ // If debugging is enabled, catch invalid strings with
+ // an assertion. However, when not debugging, use the
+ // byte count as the fallback width. This shouldn't
+ // ever happen unless there is a bad string in the
+ // translations, but in such case I guess it's better
+ // to try to print something useful instead of failing
+ // completely.
+ assert(widths[i] != (size_t)-1);
+ if (widths[i] == (size_t)-1)
+ widths[i] = lens[i];
+
+ if (widths[i] > width_max)
+ width_max = widths[i];
+ }
+
+ // Calculate the field width for printf("%*s") so that the strings
+ // will use width_max columns on a terminal.
+ for (unsigned i = 0; i < ARRAY_SIZE(colon_strs); ++i)
+ colon_strs_fw[i] = (int)(lens[i] + width_max - widths[i]);
+
+ return;
+}
+
+
+/// Initialize headings[].
+static void
+init_headings(void)
+{
+ // Before going through the heading strings themselves, treat
+ // the Check heading specially: Look at the widths of the various
+ // check names and increase the width of the Check column if needed.
+ // The width of the heading name "Check" will then be handled normally
+ // with other heading names in the second loop in this function.
+ for (unsigned i = 0; i < ARRAY_SIZE(check_names); ++i) {
+ size_t len;
+ size_t w = tuklib_mbstr_width(_(check_names[i]), &len);
+
+ // Error handling like in init_colon_strs().
+ assert(w != (size_t)-1);
+ if (w == (size_t)-1)
+ w = len;
+
+ // If the translated string is wider than the minimum width
+ // set at compile time, increase the width.
+ if ((size_t)(headings[HEADING_CHECK].columns) < w)
+ headings[HEADING_CHECK].columns = (int)w;
+ }
+
+ for (unsigned i = 0; i < ARRAY_SIZE(headings); ++i) {
+ size_t len;
+ size_t w = tuklib_mbstr_width(_(headings[i].str), &len);
+
+ // Error handling like in init_colon_strs().
+ assert(w != (size_t)-1);
+ if (w == (size_t)-1)
+ w = len;
+
+ // If the translated string is wider than the minimum width
+ // set at compile time, increase the width.
+ if ((size_t)(headings[i].columns) < w)
+ headings[i].columns = (int)w;
+
+ // Calculate the field width for printf("%*s") so that
+ // the string uses .columns number of columns on a terminal.
+ headings[i].fw = (int)(len + (size_t)headings[i].columns - w);
+ }
+
+ return;
+}
+
+
+/// Initialize the printf field widths that are needed to get nicely aligned
+/// output with translated strings.
+static void
+init_field_widths(void)
+{
+ init_colon_strs();
+ init_headings();
+ return;
+}
+
+
+/// Convert XZ Utils version number to a string.
+static const char *
+xz_ver_to_str(uint32_t ver)
+{
+ static char buf[32];
+
+ unsigned int major = ver / 10000000U;
+ ver -= major * 10000000U;
+
+ unsigned int minor = ver / 10000U;
+ ver -= minor * 10000U;
+
+ unsigned int patch = ver / 10U;
+ ver -= patch * 10U;
+
+ const char *stability = ver == 0 ? "alpha" : ver == 1 ? "beta" : "";
+
+ snprintf(buf, sizeof(buf), "%u.%u.%u%s",
+ major, minor, patch, stability);
+ return buf;
+}
+
+
+/// \brief Parse the Index(es) from the given .xz file
+///
+/// \param xfi Pointer to structure where the decoded information
+/// is stored.
+/// \param pair Input file
+///
+/// \return On success, false is returned. On error, true is returned.
+///
+static bool
+parse_indexes(xz_file_info *xfi, file_pair *pair)
+{
+ if (pair->src_st.st_size <= 0) {
+ message_error(_("%s: File is empty"), pair->src_name);
+ return true;
+ }
+
+ if (pair->src_st.st_size < 2 * LZMA_STREAM_HEADER_SIZE) {
+ message_error(_("%s: Too small to be a valid .xz file"),
+ pair->src_name);
+ return true;
+ }
+
+ io_buf buf;
+ lzma_stream strm = LZMA_STREAM_INIT;
+ lzma_index *idx = NULL;
+
+ lzma_ret ret = lzma_file_info_decoder(&strm, &idx,
+ hardware_memlimit_get(MODE_LIST),
+ (uint64_t)(pair->src_st.st_size));
+ if (ret != LZMA_OK) {
+ message_error(_("%s: %s"), pair->src_name, message_strm(ret));
+ return true;
+ }
+
+ while (true) {
+ if (strm.avail_in == 0) {
+ strm.next_in = buf.u8;
+ strm.avail_in = io_read(pair, &buf, IO_BUFFER_SIZE);
+ if (strm.avail_in == SIZE_MAX)
+ goto error;
+ }
+
+ ret = lzma_code(&strm, LZMA_RUN);
+
+ switch (ret) {
+ case LZMA_OK:
+ break;
+
+ case LZMA_SEEK_NEEDED:
+ // liblzma won't ask us to seek past the known size
+ // of the input file.
+ assert(strm.seek_pos
+ <= (uint64_t)(pair->src_st.st_size));
+ if (io_seek_src(pair, strm.seek_pos))
+ goto error;
+
+ // avail_in must be zero so that we will read new
+ // input.
+ strm.avail_in = 0;
+ break;
+
+ case LZMA_STREAM_END: {
+ lzma_end(&strm);
+ xfi->idx = idx;
+
+ // Calculate xfi->stream_padding.
+ lzma_index_iter iter;
+ lzma_index_iter_init(&iter, xfi->idx);
+ while (!lzma_index_iter_next(&iter,
+ LZMA_INDEX_ITER_STREAM))
+ xfi->stream_padding += iter.stream.padding;
+
+ return false;
+ }
+
+ default:
+ message_error(_("%s: %s"), pair->src_name,
+ message_strm(ret));
+
+ // If the error was too low memory usage limit,
+ // show also how much memory would have been needed.
+ if (ret == LZMA_MEMLIMIT_ERROR)
+ message_mem_needed(V_ERROR,
+ lzma_memusage(&strm));
+
+ goto error;
+ }
+ }
+
+error:
+ lzma_end(&strm);
+ return true;
+}
+
+
+/// \brief Parse the Block Header
+///
+/// The result is stored into *bhi. The caller takes care of initializing it.
+///
+/// \return False on success, true on error.
+static bool
+parse_block_header(file_pair *pair, const lzma_index_iter *iter,
+ block_header_info *bhi, xz_file_info *xfi)
+{
+#if IO_BUFFER_SIZE < LZMA_BLOCK_HEADER_SIZE_MAX
+# error IO_BUFFER_SIZE < LZMA_BLOCK_HEADER_SIZE_MAX
+#endif
+
+ // Get the whole Block Header with one read, but don't read past
+ // the end of the Block (or even its Check field).
+ const uint32_t size = my_min(iter->block.total_size
+ - lzma_check_size(iter->stream.flags->check),
+ LZMA_BLOCK_HEADER_SIZE_MAX);
+ io_buf buf;
+ if (io_pread(pair, &buf, size, iter->block.compressed_file_offset))
+ return true;
+
+ // Zero would mean Index Indicator and thus not a valid Block.
+ if (buf.u8[0] == 0)
+ goto data_error;
+
+ // Initialize the block structure and decode Block Header Size.
+ lzma_filter filters[LZMA_FILTERS_MAX + 1];
+ lzma_block block;
+ block.version = 0;
+ block.check = iter->stream.flags->check;
+ block.filters = filters;
+
+ block.header_size = lzma_block_header_size_decode(buf.u8[0]);
+ if (block.header_size > size)
+ goto data_error;
+
+ // Decode the Block Header.
+ switch (lzma_block_header_decode(&block, NULL, buf.u8)) {
+ case LZMA_OK:
+ break;
+
+ case LZMA_OPTIONS_ERROR:
+ message_error(_("%s: %s"), pair->src_name,
+ message_strm(LZMA_OPTIONS_ERROR));
+ return true;
+
+ case LZMA_DATA_ERROR:
+ goto data_error;
+
+ default:
+ message_bug();
+ }
+
+ // Check the Block Flags. These must be done before calling
+ // lzma_block_compressed_size(), because it overwrites
+ // block.compressed_size.
+ //
+ // NOTE: If you add new characters here, update the minimum number of
+ // columns in headings[HEADING_HEADERFLAGS] to match the number of
+ // characters used here.
+ bhi->flags[0] = block.compressed_size != LZMA_VLI_UNKNOWN
+ ? 'c' : '-';
+ bhi->flags[1] = block.uncompressed_size != LZMA_VLI_UNKNOWN
+ ? 'u' : '-';
+ bhi->flags[2] = '\0';
+
+ // Collect information if all Blocks have both Compressed Size
+ // and Uncompressed Size fields. They can be useful e.g. for
+ // multi-threaded decompression so it can be useful to know it.
+ xfi->all_have_sizes &= block.compressed_size != LZMA_VLI_UNKNOWN
+ && block.uncompressed_size != LZMA_VLI_UNKNOWN;
+
+ // Validate or set block.compressed_size.
+ switch (lzma_block_compressed_size(&block,
+ iter->block.unpadded_size)) {
+ case LZMA_OK:
+ // Validate also block.uncompressed_size if it is present.
+ // If it isn't present, there's no need to set it since
+ // we aren't going to actually decompress the Block; if
+ // we were decompressing, then we should set it so that
+ // the Block decoder could validate the Uncompressed Size
+ // that was stored in the Index.
+ if (block.uncompressed_size == LZMA_VLI_UNKNOWN
+ || block.uncompressed_size
+ == iter->block.uncompressed_size)
+ break;
+
+ // If the above fails, the file is corrupt so
+ // LZMA_DATA_ERROR is a good error code.
+
+ // Fall through
+
+ case LZMA_DATA_ERROR:
+ // Free the memory allocated by lzma_block_header_decode().
+ lzma_filters_free(filters, NULL);
+ goto data_error;
+
+ default:
+ message_bug();
+ }
+
+ // Copy the known sizes.
+ bhi->header_size = block.header_size;
+ bhi->compressed_size = block.compressed_size;
+
+ // Calculate the decoder memory usage and update the maximum
+ // memory usage of this Block.
+ bhi->memusage = lzma_raw_decoder_memusage(filters);
+ if (xfi->memusage_max < bhi->memusage)
+ xfi->memusage_max = bhi->memusage;
+
+ // Determine the minimum XZ Utils version that supports this Block.
+ //
+ // - ARM64 filter needs 5.4.0.
+ //
+ // - 5.0.0 doesn't support empty LZMA2 streams and thus empty
+ // Blocks that use LZMA2. This decoder bug was fixed in 5.0.2.
+ if (xfi->min_version < 50040002U) {
+ for (size_t i = 0; filters[i].id != LZMA_VLI_UNKNOWN; ++i) {
+ if (filters[i].id == LZMA_FILTER_ARM64) {
+ xfi->min_version = 50040002U;
+ break;
+ }
+ }
+ }
+
+ if (xfi->min_version < 50000022U) {
+ size_t i = 0;
+ while (filters[i + 1].id != LZMA_VLI_UNKNOWN)
+ ++i;
+
+ if (filters[i].id == LZMA_FILTER_LZMA2
+ && iter->block.uncompressed_size == 0)
+ xfi->min_version = 50000022U;
+ }
+
+ // Convert the filter chain to human readable form.
+ const lzma_ret str_ret = lzma_str_from_filters(
+ &bhi->filter_chain, filters,
+ LZMA_STR_DECODER | LZMA_STR_GETOPT_LONG, NULL);
+
+ // Free the memory allocated by lzma_block_header_decode().
+ lzma_filters_free(filters, NULL);
+
+ // Check if the stringification succeeded.
+ if (str_ret != LZMA_OK) {
+ message_error(_("%s: %s"), pair->src_name,
+ message_strm(str_ret));
+ return true;
+ }
+
+ return false;
+
+data_error:
+ // Show the error message.
+ message_error(_("%s: %s"), pair->src_name,
+ message_strm(LZMA_DATA_ERROR));
+ return true;
+}
+
+
+/// \brief Parse the Check field and put it into check_value[]
+///
+/// \return False on success, true on error.
+static bool
+parse_check_value(file_pair *pair, const lzma_index_iter *iter)
+{
+ // Don't read anything from the file if there is no integrity Check.
+ if (iter->stream.flags->check == LZMA_CHECK_NONE) {
+ snprintf(check_value, sizeof(check_value), "---");
+ return false;
+ }
+
+ // Locate and read the Check field.
+ const uint32_t size = lzma_check_size(iter->stream.flags->check);
+ const uint64_t offset = iter->block.compressed_file_offset
+ + iter->block.total_size - size;
+ io_buf buf;
+ if (io_pread(pair, &buf, size, offset))
+ return true;
+
+ // CRC32 and CRC64 are in little endian. Guess that all the future
+ // 32-bit and 64-bit Check values are little endian too. It shouldn't
+ // be a too big problem if this guess is wrong.
+ if (size == 4)
+ snprintf(check_value, sizeof(check_value),
+ "%08" PRIx32, conv32le(buf.u32[0]));
+ else if (size == 8)
+ snprintf(check_value, sizeof(check_value),
+ "%016" PRIx64, conv64le(buf.u64[0]));
+ else
+ for (size_t i = 0; i < size; ++i)
+ snprintf(check_value + i * 2, 3, "%02x", buf.u8[i]);
+
+ return false;
+}
+
+
+/// \brief Parse detailed information about a Block
+///
+/// Since this requires seek(s), listing information about all Blocks can
+/// be slow.
+///
+/// \param pair Input file
+/// \param iter Location of the Block whose Check value should
+/// be printed.
+/// \param bhi Pointer to structure where to store the information
+/// about the Block Header field.
+///
+/// \return False on success, true on error. If an error occurs,
+/// the error message is printed too so the caller doesn't
+/// need to worry about that.
+static bool
+parse_details(file_pair *pair, const lzma_index_iter *iter,
+ block_header_info *bhi, xz_file_info *xfi)
+{
+ if (parse_block_header(pair, iter, bhi, xfi))
+ return true;
+
+ if (parse_check_value(pair, iter))
+ return true;
+
+ return false;
+}
+
+
+/// \brief Get the compression ratio
+///
+/// This has slightly different format than that is used in message.c.
+static const char *
+get_ratio(uint64_t compressed_size, uint64_t uncompressed_size)
+{
+ if (uncompressed_size == 0)
+ return "---";
+
+ const double ratio = (double)(compressed_size)
+ / (double)(uncompressed_size);
+ if (ratio > 9.999)
+ return "---";
+
+ static char buf[16];
+ snprintf(buf, sizeof(buf), "%.3f", ratio);
+ return buf;
+}
+
+
+/// \brief Get a comma-separated list of Check names
+///
+/// The check names are translated with gettext except when in robot mode.
+///
+/// \param buf Buffer to hold the resulting string
+/// \param checks Bit mask of Checks to print
+/// \param space_after_comma
+/// It's better to not use spaces in table-like listings,
+/// but in more verbose formats a space after a comma
+/// is good for readability.
+static void
+get_check_names(char buf[CHECKS_STR_SIZE],
+ uint32_t checks, bool space_after_comma)
+{
+ // If we get called when there are no Checks to print, set checks
+ // to 1 so that we print "None". This can happen in the robot mode
+ // when printing the totals line if there are no valid input files.
+ if (checks == 0)
+ checks = 1;
+
+ char *pos = buf;
+ size_t left = CHECKS_STR_SIZE;
+
+ const char *sep = space_after_comma ? ", " : ",";
+ bool comma = false;
+
+ for (size_t i = 0; i <= LZMA_CHECK_ID_MAX; ++i) {
+ if (checks & (UINT32_C(1) << i)) {
+ my_snprintf(&pos, &left, "%s%s",
+ comma ? sep : "",
+ opt_robot ? check_names[i]
+ : _(check_names[i]));
+ comma = true;
+ }
+ }
+
+ return;
+}
+
+
+static bool
+print_info_basic(const xz_file_info *xfi, file_pair *pair)
+{
+ static bool headings_displayed = false;
+ if (!headings_displayed) {
+ headings_displayed = true;
+ // TRANSLATORS: These are column headings. From Strms (Streams)
+ // to Ratio, the columns are right aligned. Check and Filename
+ // are left aligned. If you need longer words, it's OK to
+ // use two lines here. Test with "xz -l foo.xz".
+ puts(_("Strms Blocks Compressed Uncompressed Ratio "
+ "Check Filename"));
+ }
+
+ char checks[CHECKS_STR_SIZE];
+ get_check_names(checks, lzma_index_checks(xfi->idx), false);
+
+ const char *cols[7] = {
+ uint64_to_str(lzma_index_stream_count(xfi->idx), 0),
+ uint64_to_str(lzma_index_block_count(xfi->idx), 1),
+ uint64_to_nicestr(lzma_index_file_size(xfi->idx),
+ NICESTR_B, NICESTR_TIB, false, 2),
+ uint64_to_nicestr(lzma_index_uncompressed_size(xfi->idx),
+ NICESTR_B, NICESTR_TIB, false, 3),
+ get_ratio(lzma_index_file_size(xfi->idx),
+ lzma_index_uncompressed_size(xfi->idx)),
+ checks,
+ pair->src_name,
+ };
+ printf("%*s %*s %*s %*s %*s %-*s %s\n",
+ tuklib_mbstr_fw(cols[0], 5), cols[0],
+ tuklib_mbstr_fw(cols[1], 7), cols[1],
+ tuklib_mbstr_fw(cols[2], 11), cols[2],
+ tuklib_mbstr_fw(cols[3], 11), cols[3],
+ tuklib_mbstr_fw(cols[4], 5), cols[4],
+ tuklib_mbstr_fw(cols[5], 7), cols[5],
+ cols[6]);
+
+ return false;
+}
+
+
+static void
+print_adv_helper(uint64_t stream_count, uint64_t block_count,
+ uint64_t compressed_size, uint64_t uncompressed_size,
+ uint32_t checks, uint64_t stream_padding)
+{
+ char checks_str[CHECKS_STR_SIZE];
+ get_check_names(checks_str, checks, true);
+
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_STREAMS),
+ uint64_to_str(stream_count, 0));
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_BLOCKS),
+ uint64_to_str(block_count, 0));
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_COMPRESSED_SIZE),
+ uint64_to_nicestr(compressed_size,
+ NICESTR_B, NICESTR_TIB, true, 0));
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_UNCOMPRESSED_SIZE),
+ uint64_to_nicestr(uncompressed_size,
+ NICESTR_B, NICESTR_TIB, true, 0));
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_RATIO),
+ get_ratio(compressed_size, uncompressed_size));
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_CHECK), checks_str);
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_STREAM_PADDING),
+ uint64_to_nicestr(stream_padding,
+ NICESTR_B, NICESTR_TIB, true, 0));
+ return;
+}
+
+
+static bool
+print_info_adv(xz_file_info *xfi, file_pair *pair)
+{
+ // Print the overall information.
+ print_adv_helper(lzma_index_stream_count(xfi->idx),
+ lzma_index_block_count(xfi->idx),
+ lzma_index_file_size(xfi->idx),
+ lzma_index_uncompressed_size(xfi->idx),
+ lzma_index_checks(xfi->idx),
+ xfi->stream_padding);
+
+ // Size of the biggest Check. This is used to calculate the width
+ // of the CheckVal field. The table would get insanely wide if
+ // we always reserved space for 64-byte Check (128 chars as hex).
+ uint32_t check_max = 0;
+
+ // Print information about the Streams.
+ //
+ // All except Check are right aligned; Check is left aligned.
+ // Test with "xz -lv foo.xz".
+ printf(" %s\n %*s %*s %*s %*s %*s %*s %*s %-*s %*s\n",
+ _(colon_strs[COLON_STR_STREAMS]),
+ HEADING_STR(HEADING_STREAM),
+ HEADING_STR(HEADING_BLOCKS),
+ HEADING_STR(HEADING_COMPOFFSET),
+ HEADING_STR(HEADING_UNCOMPOFFSET),
+ HEADING_STR(HEADING_COMPSIZE),
+ HEADING_STR(HEADING_UNCOMPSIZE),
+ HEADING_STR(HEADING_RATIO),
+ HEADING_STR(HEADING_CHECK),
+ HEADING_STR(HEADING_PADDING));
+
+ lzma_index_iter iter;
+ lzma_index_iter_init(&iter, xfi->idx);
+
+ while (!lzma_index_iter_next(&iter, LZMA_INDEX_ITER_STREAM)) {
+ const char *cols1[4] = {
+ uint64_to_str(iter.stream.number, 0),
+ uint64_to_str(iter.stream.block_count, 1),
+ uint64_to_str(iter.stream.compressed_offset, 2),
+ uint64_to_str(iter.stream.uncompressed_offset, 3),
+ };
+ printf(" %*s %*s %*s %*s ",
+ tuklib_mbstr_fw(cols1[0],
+ headings[HEADING_STREAM].columns),
+ cols1[0],
+ tuklib_mbstr_fw(cols1[1],
+ headings[HEADING_BLOCKS].columns),
+ cols1[1],
+ tuklib_mbstr_fw(cols1[2],
+ headings[HEADING_COMPOFFSET].columns),
+ cols1[2],
+ tuklib_mbstr_fw(cols1[3],
+ headings[HEADING_UNCOMPOFFSET].columns),
+ cols1[3]);
+
+ const char *cols2[5] = {
+ uint64_to_str(iter.stream.compressed_size, 0),
+ uint64_to_str(iter.stream.uncompressed_size, 1),
+ get_ratio(iter.stream.compressed_size,
+ iter.stream.uncompressed_size),
+ _(check_names[iter.stream.flags->check]),
+ uint64_to_str(iter.stream.padding, 2),
+ };
+ printf("%*s %*s %*s %-*s %*s\n",
+ tuklib_mbstr_fw(cols2[0],
+ headings[HEADING_COMPSIZE].columns),
+ cols2[0],
+ tuklib_mbstr_fw(cols2[1],
+ headings[HEADING_UNCOMPSIZE].columns),
+ cols2[1],
+ tuklib_mbstr_fw(cols2[2],
+ headings[HEADING_RATIO].columns),
+ cols2[2],
+ tuklib_mbstr_fw(cols2[3],
+ headings[HEADING_CHECK].columns),
+ cols2[3],
+ tuklib_mbstr_fw(cols2[4],
+ headings[HEADING_PADDING].columns),
+ cols2[4]);
+
+ // Update the maximum Check size.
+ if (lzma_check_size(iter.stream.flags->check) > check_max)
+ check_max = lzma_check_size(iter.stream.flags->check);
+ }
+
+ // Cache the verbosity level to a local variable.
+ const bool detailed = message_verbosity_get() >= V_DEBUG;
+
+ // Print information about the Blocks but only if there is
+ // at least one Block.
+ if (lzma_index_block_count(xfi->idx) > 0) {
+ // Calculate the width of the CheckVal column. This can be
+ // used as is as the field width for printf() when printing
+ // the actual check value as it is hexadecimal. However, to
+ // print the column heading, further calculation is needed
+ // to handle a translated string (it's done a few lines later).
+ assert(check_max <= LZMA_CHECK_SIZE_MAX);
+ const int checkval_width = my_max(
+ headings[HEADING_CHECKVAL].columns,
+ (int)(2 * check_max));
+
+ // All except Check are right aligned; Check is left aligned.
+ printf(" %s\n %*s %*s %*s %*s %*s %*s %*s %-*s",
+ _(colon_strs[COLON_STR_BLOCKS]),
+ HEADING_STR(HEADING_STREAM),
+ HEADING_STR(HEADING_BLOCK),
+ HEADING_STR(HEADING_COMPOFFSET),
+ HEADING_STR(HEADING_UNCOMPOFFSET),
+ HEADING_STR(HEADING_TOTALSIZE),
+ HEADING_STR(HEADING_UNCOMPSIZE),
+ HEADING_STR(HEADING_RATIO),
+ detailed ? headings[HEADING_CHECK].fw : 1,
+ _(headings[HEADING_CHECK].str));
+
+ if (detailed) {
+ // CheckVal (Check value), Flags, and Filters are
+ // left aligned. Block Header Size, CompSize, and
+ // MemUsage are right aligned. Test with
+ // "xz -lvv foo.xz".
+ printf(" %-*s %*s %-*s %*s %*s %s",
+ headings[HEADING_CHECKVAL].fw
+ + checkval_width
+ - headings[HEADING_CHECKVAL].columns,
+ _(headings[HEADING_CHECKVAL].str),
+ HEADING_STR(HEADING_HEADERSIZE),
+ HEADING_STR(HEADING_HEADERFLAGS),
+ HEADING_STR(HEADING_COMPSIZE),
+ HEADING_STR(HEADING_MEMUSAGE),
+ _(headings[HEADING_FILTERS].str));
+ }
+
+ putchar('\n');
+
+ lzma_index_iter_init(&iter, xfi->idx);
+
+ // Iterate over the Blocks.
+ while (!lzma_index_iter_next(&iter, LZMA_INDEX_ITER_BLOCK)) {
+ // If in detailed mode, collect the information from
+ // Block Header before starting to print the next line.
+ block_header_info bhi = BLOCK_HEADER_INFO_INIT;
+ if (detailed && parse_details(pair, &iter, &bhi, xfi))
+ return true;
+
+ const char *cols1[4] = {
+ uint64_to_str(iter.stream.number, 0),
+ uint64_to_str(
+ iter.block.number_in_stream, 1),
+ uint64_to_str(
+ iter.block.compressed_file_offset, 2),
+ uint64_to_str(
+ iter.block.uncompressed_file_offset, 3)
+ };
+ printf(" %*s %*s %*s %*s ",
+ tuklib_mbstr_fw(cols1[0],
+ headings[HEADING_STREAM].columns),
+ cols1[0],
+ tuklib_mbstr_fw(cols1[1],
+ headings[HEADING_BLOCK].columns),
+ cols1[1],
+ tuklib_mbstr_fw(cols1[2],
+ headings[HEADING_COMPOFFSET].columns),
+ cols1[2],
+ tuklib_mbstr_fw(cols1[3], headings[
+ HEADING_UNCOMPOFFSET].columns),
+ cols1[3]);
+
+ const char *cols2[4] = {
+ uint64_to_str(iter.block.total_size, 0),
+ uint64_to_str(iter.block.uncompressed_size,
+ 1),
+ get_ratio(iter.block.total_size,
+ iter.block.uncompressed_size),
+ _(check_names[iter.stream.flags->check])
+ };
+ printf("%*s %*s %*s %-*s",
+ tuklib_mbstr_fw(cols2[0],
+ headings[HEADING_TOTALSIZE].columns),
+ cols2[0],
+ tuklib_mbstr_fw(cols2[1],
+ headings[HEADING_UNCOMPSIZE].columns),
+ cols2[1],
+ tuklib_mbstr_fw(cols2[2],
+ headings[HEADING_RATIO].columns),
+ cols2[2],
+ tuklib_mbstr_fw(cols2[3], detailed
+ ? headings[HEADING_CHECK].columns : 1),
+ cols2[3]);
+
+ if (detailed) {
+ const lzma_vli compressed_size
+ = iter.block.unpadded_size
+ - bhi.header_size
+ - lzma_check_size(
+ iter.stream.flags->check);
+
+ const char *cols3[6] = {
+ check_value,
+ uint64_to_str(bhi.header_size, 0),
+ bhi.flags,
+ uint64_to_str(compressed_size, 1),
+ uint64_to_str(
+ round_up_to_mib(bhi.memusage),
+ 2),
+ bhi.filter_chain
+ };
+ // Show MiB for memory usage, because it
+ // is the only size which is not in bytes.
+ printf(" %-*s %*s %-*s %*s %*s MiB %s",
+ checkval_width, cols3[0],
+ tuklib_mbstr_fw(cols3[1], headings[
+ HEADING_HEADERSIZE].columns),
+ cols3[1],
+ tuklib_mbstr_fw(cols3[2], headings[
+ HEADING_HEADERFLAGS].columns),
+ cols3[2],
+ tuklib_mbstr_fw(cols3[3], headings[
+ HEADING_COMPSIZE].columns),
+ cols3[3],
+ tuklib_mbstr_fw(cols3[4], headings[
+ HEADING_MEMUSAGE].columns - 4),
+ cols3[4],
+ cols3[5]);
+ }
+
+ putchar('\n');
+ block_header_info_end(&bhi);
+ }
+ }
+
+ if (detailed) {
+ printf(" %-*s %s MiB\n", COLON_STR(COLON_STR_MEMORY_NEEDED),
+ uint64_to_str(
+ round_up_to_mib(xfi->memusage_max), 0));
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_SIZES_IN_HEADERS),
+ xfi->all_have_sizes ? _("Yes") : _("No"));
+ //printf(" %-*s %s\n", COLON_STR(COLON_STR_MINIMUM_XZ_VERSION),
+ printf(_(" Minimum XZ Utils version: %s\n"),
+ xz_ver_to_str(xfi->min_version));
+ }
+
+ return false;
+}
+
+
+static bool
+print_info_robot(xz_file_info *xfi, file_pair *pair)
+{
+ char checks[CHECKS_STR_SIZE];
+ get_check_names(checks, lzma_index_checks(xfi->idx), false);
+
+ printf("name\t%s\n", pair->src_name);
+
+ printf("file\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%s\t%s\t%" PRIu64 "\n",
+ lzma_index_stream_count(xfi->idx),
+ lzma_index_block_count(xfi->idx),
+ lzma_index_file_size(xfi->idx),
+ lzma_index_uncompressed_size(xfi->idx),
+ get_ratio(lzma_index_file_size(xfi->idx),
+ lzma_index_uncompressed_size(xfi->idx)),
+ checks,
+ xfi->stream_padding);
+
+ if (message_verbosity_get() >= V_VERBOSE) {
+ lzma_index_iter iter;
+ lzma_index_iter_init(&iter, xfi->idx);
+
+ while (!lzma_index_iter_next(&iter, LZMA_INDEX_ITER_STREAM))
+ printf("stream\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%s\t%s\t%" PRIu64 "\n",
+ iter.stream.number,
+ iter.stream.block_count,
+ iter.stream.compressed_offset,
+ iter.stream.uncompressed_offset,
+ iter.stream.compressed_size,
+ iter.stream.uncompressed_size,
+ get_ratio(iter.stream.compressed_size,
+ iter.stream.uncompressed_size),
+ check_names[iter.stream.flags->check],
+ iter.stream.padding);
+
+ lzma_index_iter_rewind(&iter);
+
+ while (!lzma_index_iter_next(&iter, LZMA_INDEX_ITER_BLOCK)) {
+ block_header_info bhi = BLOCK_HEADER_INFO_INIT;
+ if (message_verbosity_get() >= V_DEBUG
+ && parse_details(
+ pair, &iter, &bhi, xfi))
+ return true;
+
+ printf("block\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%" PRIu64 "\t%" PRIu64
+ "\t%" PRIu64 "\t%" PRIu64 "\t%s\t%s",
+ iter.stream.number,
+ iter.block.number_in_stream,
+ iter.block.number_in_file,
+ iter.block.compressed_file_offset,
+ iter.block.uncompressed_file_offset,
+ iter.block.total_size,
+ iter.block.uncompressed_size,
+ get_ratio(iter.block.total_size,
+ iter.block.uncompressed_size),
+ check_names[iter.stream.flags->check]);
+
+ if (message_verbosity_get() >= V_DEBUG)
+ printf("\t%s\t%" PRIu32 "\t%s\t%" PRIu64
+ "\t%" PRIu64 "\t%s",
+ check_value,
+ bhi.header_size,
+ bhi.flags,
+ bhi.compressed_size,
+ bhi.memusage,
+ bhi.filter_chain);
+
+ putchar('\n');
+ block_header_info_end(&bhi);
+ }
+ }
+
+ if (message_verbosity_get() >= V_DEBUG)
+ printf("summary\t%" PRIu64 "\t%s\t%" PRIu32 "\n",
+ xfi->memusage_max,
+ xfi->all_have_sizes ? "yes" : "no",
+ xfi->min_version);
+
+ return false;
+}
+
+
+static void
+update_totals(const xz_file_info *xfi)
+{
+ // TODO: Integer overflow checks
+ ++totals.files;
+ totals.streams += lzma_index_stream_count(xfi->idx);
+ totals.blocks += lzma_index_block_count(xfi->idx);
+ totals.compressed_size += lzma_index_file_size(xfi->idx);
+ totals.uncompressed_size += lzma_index_uncompressed_size(xfi->idx);
+ totals.stream_padding += xfi->stream_padding;
+ totals.checks |= lzma_index_checks(xfi->idx);
+
+ if (totals.memusage_max < xfi->memusage_max)
+ totals.memusage_max = xfi->memusage_max;
+
+ if (totals.min_version < xfi->min_version)
+ totals.min_version = xfi->min_version;
+
+ totals.all_have_sizes &= xfi->all_have_sizes;
+
+ return;
+}
+
+
+static void
+print_totals_basic(void)
+{
+ // Print a separator line.
+ char line[80];
+ memset(line, '-', sizeof(line));
+ line[sizeof(line) - 1] = '\0';
+ puts(line);
+
+ // Get the check names.
+ char checks[CHECKS_STR_SIZE];
+ get_check_names(checks, totals.checks, false);
+
+ // Print the totals except the file count, which needs
+ // special handling.
+ printf("%5s %7s %11s %11s %5s %-7s ",
+ uint64_to_str(totals.streams, 0),
+ uint64_to_str(totals.blocks, 1),
+ uint64_to_nicestr(totals.compressed_size,
+ NICESTR_B, NICESTR_TIB, false, 2),
+ uint64_to_nicestr(totals.uncompressed_size,
+ NICESTR_B, NICESTR_TIB, false, 3),
+ get_ratio(totals.compressed_size,
+ totals.uncompressed_size),
+ checks);
+
+ // Since we print totals only when there are at least two files,
+ // the English message will always use "%s files". But some other
+ // languages need different forms for different plurals so we
+ // have to translate this with ngettext().
+ //
+ // TRANSLATORS: %s is an integer. Only the plural form of this
+ // message is used (e.g. "2 files"). Test with "xz -l foo.xz bar.xz".
+ printf(ngettext("%s file\n", "%s files\n",
+ totals.files <= ULONG_MAX ? totals.files
+ : (totals.files % 1000000) + 1000000),
+ uint64_to_str(totals.files, 0));
+
+ return;
+}
+
+
+static void
+print_totals_adv(void)
+{
+ putchar('\n');
+ puts(_("Totals:"));
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_NUMBER_OF_FILES),
+ uint64_to_str(totals.files, 0));
+ print_adv_helper(totals.streams, totals.blocks,
+ totals.compressed_size, totals.uncompressed_size,
+ totals.checks, totals.stream_padding);
+
+ if (message_verbosity_get() >= V_DEBUG) {
+ printf(" %-*s %s MiB\n", COLON_STR(COLON_STR_MEMORY_NEEDED),
+ uint64_to_str(
+ round_up_to_mib(totals.memusage_max), 0));
+ printf(" %-*s %s\n", COLON_STR(COLON_STR_SIZES_IN_HEADERS),
+ totals.all_have_sizes ? _("Yes") : _("No"));
+ //printf(" %-*s %s\n", COLON_STR(COLON_STR_MINIMUM_XZ_VERSION),
+ printf(_(" Minimum XZ Utils version: %s\n"),
+ xz_ver_to_str(totals.min_version));
+ }
+
+ return;
+}
+
+
+static void
+print_totals_robot(void)
+{
+ char checks[CHECKS_STR_SIZE];
+ get_check_names(checks, totals.checks, false);
+
+ printf("totals\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%s\t%s\t%" PRIu64 "\t%" PRIu64,
+ totals.streams,
+ totals.blocks,
+ totals.compressed_size,
+ totals.uncompressed_size,
+ get_ratio(totals.compressed_size,
+ totals.uncompressed_size),
+ checks,
+ totals.stream_padding,
+ totals.files);
+
+ if (message_verbosity_get() >= V_DEBUG)
+ printf("\t%" PRIu64 "\t%s\t%" PRIu32,
+ totals.memusage_max,
+ totals.all_have_sizes ? "yes" : "no",
+ totals.min_version);
+
+ putchar('\n');
+
+ return;
+}
+
+
+extern void
+list_totals(void)
+{
+ if (opt_robot) {
+ // Always print totals in --robot mode. It can be convenient
+ // in some cases and doesn't complicate usage of the
+ // single-file case much.
+ print_totals_robot();
+
+ } else if (totals.files > 1) {
+ // For non-robot mode, totals are printed only if there
+ // is more than one file.
+ if (message_verbosity_get() <= V_WARNING)
+ print_totals_basic();
+ else
+ print_totals_adv();
+ }
+
+ return;
+}
+
+
+extern void
+list_file(const char *filename)
+{
+ if (opt_format != FORMAT_XZ && opt_format != FORMAT_AUTO)
+ message_fatal(_("--list works only on .xz files "
+ "(--format=xz or --format=auto)"));
+
+ message_filename(filename);
+
+ if (filename == stdin_filename) {
+ message_error(_("--list does not support reading from "
+ "standard input"));
+ return;
+ }
+
+ init_field_widths();
+
+ // Unset opt_stdout so that io_open_src() won't accept special files.
+ // Set opt_force so that io_open_src() will follow symlinks.
+ opt_stdout = false;
+ opt_force = true;
+ file_pair *pair = io_open_src(filename);
+ if (pair == NULL)
+ return;
+
+ xz_file_info xfi = XZ_FILE_INFO_INIT;
+ if (!parse_indexes(&xfi, pair)) {
+ bool fail;
+
+ // We have three main modes:
+ // - --robot, which has submodes if --verbose is specified
+ // once or twice
+ // - Normal --list without --verbose
+ // - --list with one or two --verbose
+ if (opt_robot)
+ fail = print_info_robot(&xfi, pair);
+ else if (message_verbosity_get() <= V_WARNING)
+ fail = print_info_basic(&xfi, pair);
+ else
+ fail = print_info_adv(&xfi, pair);
+
+ // Update the totals that are displayed after all
+ // the individual files have been listed. Don't count
+ // broken files.
+ if (!fail)
+ update_totals(&xfi);
+
+ lzma_index_end(xfi.idx, NULL);
+ }
+
+ io_close(pair, false);
+ return;
+}
diff --git a/src/xz/list.h b/src/xz/list.h
new file mode 100644
index 0000000..a4c6ec7
--- /dev/null
+++ b/src/xz/list.h
@@ -0,0 +1,18 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file list.h
+/// \brief List information about .xz files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// \brief List information about the given .xz file
+extern void list_file(const char *filename);
+
+
+/// \brief Show the totals after all files have been listed
+extern void list_totals(void);
diff --git a/src/xz/main.c b/src/xz/main.c
new file mode 100644
index 0000000..c9c3dec
--- /dev/null
+++ b/src/xz/main.c
@@ -0,0 +1,344 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file main.c
+/// \brief main()
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+#include <ctype.h>
+
+/// Exit status to use. This can be changed with set_exit_status().
+static enum exit_status_type exit_status = E_SUCCESS;
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+/// exit_status has to be protected with a critical section due to
+/// how "signal handling" is done on Windows. See signals.c for details.
+static CRITICAL_SECTION exit_status_cs;
+#endif
+
+/// True if --no-warn is specified. When this is true, we don't set
+/// the exit status to E_WARNING when something worth a warning happens.
+static bool no_warn = false;
+
+
+extern void
+set_exit_status(enum exit_status_type new_status)
+{
+ assert(new_status == E_WARNING || new_status == E_ERROR);
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ EnterCriticalSection(&exit_status_cs);
+#endif
+
+ if (exit_status != E_ERROR)
+ exit_status = new_status;
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ LeaveCriticalSection(&exit_status_cs);
+#endif
+
+ return;
+}
+
+
+extern void
+set_exit_no_warn(void)
+{
+ no_warn = true;
+ return;
+}
+
+
+static const char *
+read_name(const args_info *args)
+{
+ // FIXME: Maybe we should have some kind of memory usage limit here
+ // like the tool has for the actual compression and decompression.
+ // Giving some huge text file with --files0 makes us to read the
+ // whole file in RAM.
+ static char *name = NULL;
+ static size_t size = 256;
+
+ // Allocate the initial buffer. This is never freed, since after it
+ // is no longer needed, the program exits very soon. It is safe to
+ // use xmalloc() and xrealloc() in this function, because while
+ // executing this function, no files are open for writing, and thus
+ // there's no need to cleanup anything before exiting.
+ if (name == NULL)
+ name = xmalloc(size);
+
+ // Write position in name
+ size_t pos = 0;
+
+ // Read one character at a time into name.
+ while (!user_abort) {
+ const int c = fgetc(args->files_file);
+
+ if (ferror(args->files_file)) {
+ // Take care of EINTR since we have established
+ // the signal handlers already.
+ if (errno == EINTR)
+ continue;
+
+ message_error(_("%s: Error reading filenames: %s"),
+ args->files_name, strerror(errno));
+ return NULL;
+ }
+
+ if (feof(args->files_file)) {
+ if (pos != 0)
+ message_error(_("%s: Unexpected end of input "
+ "when reading filenames"),
+ args->files_name);
+
+ return NULL;
+ }
+
+ if (c == args->files_delim) {
+ // We allow consecutive newline (--files) or '\0'
+ // characters (--files0), and ignore such empty
+ // filenames.
+ if (pos == 0)
+ continue;
+
+ // A non-empty name was read. Terminate it with '\0'
+ // and return it.
+ name[pos] = '\0';
+ return name;
+ }
+
+ if (c == '\0') {
+ // A null character was found when using --files,
+ // which expects plain text input separated with
+ // newlines.
+ message_error(_("%s: Null character found when "
+ "reading filenames; maybe you meant "
+ "to use `--files0' instead "
+ "of `--files'?"), args->files_name);
+ return NULL;
+ }
+
+ name[pos++] = c;
+
+ // Allocate more memory if needed. There must always be space
+ // at least for one character to allow terminating the string
+ // with '\0'.
+ if (pos == size) {
+ size *= 2;
+ name = xrealloc(name, size);
+ }
+ }
+
+ return NULL;
+}
+
+
+int
+main(int argc, char **argv)
+{
+#ifdef HAVE_PLEDGE
+ // OpenBSD's pledge(2) sandbox
+ //
+ // Unconditionally enable sandboxing with fairly relaxed promises.
+ // This is still way better than having no sandbox at all. :-)
+ // More strict promises will be made later in file_io.c if possible.
+ if (pledge("stdio rpath wpath cpath fattr", "")) {
+ // Don't translate the string or use message_fatal() as
+ // those haven't been initialized yet.
+ fprintf(stderr, "%s: Failed to enable the sandbox\n", argv[0]);
+ return E_ERROR;
+ }
+#endif
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ InitializeCriticalSection(&exit_status_cs);
+#endif
+
+ // Set up the progname variable.
+ tuklib_progname_init(argv);
+
+ // Initialize the file I/O. This makes sure that
+ // stdin, stdout, and stderr are something valid.
+ io_init();
+
+ // Set up the locale and message translations.
+ tuklib_gettext_init(PACKAGE, LOCALEDIR);
+
+ // Initialize handling of error/warning/other messages.
+ message_init();
+
+ // Set hardware-dependent default values. These can be overridden
+ // on the command line, thus this must be done before args_parse().
+ hardware_init();
+
+ // Parse the command line arguments and get an array of filenames.
+ // This doesn't return if something is wrong with the command line
+ // arguments. If there are no arguments, one filename ("-") is still
+ // returned to indicate stdin.
+ args_info args;
+ args_parse(&args, argc, argv);
+
+ if (opt_mode != MODE_LIST && opt_robot)
+ message_fatal(_("Compression and decompression with --robot "
+ "are not supported yet."));
+
+ // Tell the message handling code how many input files there are if
+ // we know it. This way the progress indicator can show it.
+ if (args.files_name != NULL)
+ message_set_files(0);
+ else
+ message_set_files(args.arg_count);
+
+ // Refuse to write compressed data to standard output if it is
+ // a terminal.
+ if (opt_mode == MODE_COMPRESS) {
+ if (opt_stdout || (args.arg_count == 1
+ && strcmp(args.arg_names[0], "-") == 0)) {
+ if (is_tty_stdout()) {
+ message_try_help();
+ tuklib_exit(E_ERROR, E_ERROR, false);
+ }
+ }
+ }
+
+ // Set up the signal handlers. We don't need these before we
+ // start the actual action and not in --list mode, so this is
+ // done after parsing the command line arguments.
+ //
+ // It's good to keep signal handlers in normal compression and
+ // decompression modes even when only writing to stdout, because
+ // we might need to restore O_APPEND flag on stdout before exiting.
+ // In --test mode, signal handlers aren't really needed, but let's
+ // keep them there for consistency with normal decompression.
+ if (opt_mode != MODE_LIST)
+ signals_init();
+
+#ifdef ENABLE_SANDBOX
+ // Set a flag that sandboxing is allowed if all these are true:
+ // - --files or --files0 wasn't used.
+ // - There is exactly one input file or we are reading from stdin.
+ // - We won't create any files: output goes to stdout or --test
+ // or --list was used. Note that --test implies opt_stdout = true
+ // but --list doesn't.
+ //
+ // This is obviously not ideal but it was easy to implement and
+ // it covers the most common use cases.
+ //
+ // TODO: Make sandboxing work for other situations too.
+ if (args.files_name == NULL && args.arg_count == 1
+ && (opt_stdout || strcmp("-", args.arg_names[0]) == 0
+ || opt_mode == MODE_LIST))
+ io_allow_sandbox();
+#endif
+
+ // coder_run() handles compression, decompression, and testing.
+ // list_file() is for --list.
+ void (*run)(const char *filename) = &coder_run;
+#ifdef HAVE_DECODERS
+ if (opt_mode == MODE_LIST)
+ run = &list_file;
+#endif
+
+ // Process the files given on the command line. Note that if no names
+ // were given, args_parse() gave us a fake "-" filename.
+ for (unsigned i = 0; i < args.arg_count && !user_abort; ++i) {
+ if (strcmp("-", args.arg_names[i]) == 0) {
+ // Processing from stdin to stdout. Check that we
+ // aren't writing compressed data to a terminal or
+ // reading it from a terminal.
+ if (opt_mode == MODE_COMPRESS) {
+ if (is_tty_stdout())
+ continue;
+ } else if (is_tty_stdin()) {
+ continue;
+ }
+
+ // It doesn't make sense to compress data from stdin
+ // if we are supposed to read filenames from stdin
+ // too (enabled with --files or --files0).
+ if (args.files_name == stdin_filename) {
+ message_error(_("Cannot read data from "
+ "standard input when "
+ "reading filenames "
+ "from standard input"));
+ continue;
+ }
+
+ // Replace the "-" with a special pointer, which is
+ // recognized by coder_run() and other things.
+ // This way error messages get a proper filename
+ // string and the code still knows that it is
+ // handling the special case of stdin.
+ args.arg_names[i] = (char *)stdin_filename;
+ }
+
+ // Do the actual compression or decompression.
+ run(args.arg_names[i]);
+ }
+
+ // If --files or --files0 was used, process the filenames from the
+ // given file or stdin. Note that here we don't consider "-" to
+ // indicate stdin like we do with the command line arguments.
+ if (args.files_name != NULL) {
+ // read_name() checks for user_abort so we don't need to
+ // check it as loop termination condition.
+ while (true) {
+ const char *name = read_name(&args);
+ if (name == NULL)
+ break;
+
+ // read_name() doesn't return empty names.
+ assert(name[0] != '\0');
+ run(name);
+ }
+
+ if (args.files_name != stdin_filename)
+ (void)fclose(args.files_file);
+ }
+
+#ifdef HAVE_DECODERS
+ // All files have now been handled. If in --list mode, display
+ // the totals before exiting. We don't have signal handlers
+ // enabled in --list mode, so we don't need to check user_abort.
+ if (opt_mode == MODE_LIST) {
+ assert(!user_abort);
+ list_totals();
+ }
+#endif
+
+#ifndef NDEBUG
+ coder_free();
+ args_free();
+#endif
+
+ // If we have got a signal, raise it to kill the program instead
+ // of calling tuklib_exit().
+ signals_exit();
+
+ // Make a local copy of exit_status to keep the Windows code
+ // thread safe. At this point it is fine if we miss the user
+ // pressing C-c and don't set the exit_status to E_ERROR on
+ // Windows.
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ EnterCriticalSection(&exit_status_cs);
+#endif
+
+ enum exit_status_type es = exit_status;
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ LeaveCriticalSection(&exit_status_cs);
+#endif
+
+ // Suppress the exit status indicating a warning if --no-warn
+ // was specified.
+ if (es == E_WARNING && no_warn)
+ es = E_SUCCESS;
+
+ tuklib_exit((int)es, E_ERROR, message_verbosity_get() != V_SILENT);
+}
diff --git a/src/xz/main.h b/src/xz/main.h
new file mode 100644
index 0000000..323f2f7
--- /dev/null
+++ b/src/xz/main.h
@@ -0,0 +1,30 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file main.h
+/// \brief Miscellaneous declarations
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// Possible exit status values. These are the same as used by gzip and bzip2.
+enum exit_status_type {
+ E_SUCCESS = 0,
+ E_ERROR = 1,
+ E_WARNING = 2,
+};
+
+
+/// Sets the exit status after a warning or error has occurred. If new_status
+/// is E_WARNING and the old exit status was already E_ERROR, the exit
+/// status is not changed.
+extern void set_exit_status(enum exit_status_type new_status);
+
+
+/// Use E_SUCCESS instead of E_WARNING if something worth a warning occurs
+/// but nothing worth an error has occurred. This is called when --no-warn
+/// is specified.
+extern void set_exit_no_warn(void);
diff --git a/src/xz/message.c b/src/xz/message.c
new file mode 100644
index 0000000..abf30ad
--- /dev/null
+++ b/src/xz/message.c
@@ -0,0 +1,1146 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file message.c
+/// \brief Printing messages
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+#include <stdarg.h>
+
+
+/// Number of the current file
+static unsigned int files_pos = 0;
+
+/// Total number of input files; zero if unknown.
+static unsigned int files_total;
+
+/// Verbosity level
+static enum message_verbosity verbosity = V_WARNING;
+
+/// Filename which we will print with the verbose messages
+static const char *filename;
+
+/// True once the a filename has been printed to stderr as part of progress
+/// message. If automatic progress updating isn't enabled, this becomes true
+/// after the first progress message has been printed due to user sending
+/// SIGINFO, SIGUSR1, or SIGALRM. Once this variable is true, we will print
+/// an empty line before the next filename to make the output more readable.
+static bool first_filename_printed = false;
+
+/// This is set to true when we have printed the current filename to stderr
+/// as part of a progress message. This variable is useful only if not
+/// updating progress automatically: if user sends many SIGINFO, SIGUSR1, or
+/// SIGALRM signals, we won't print the name of the same file multiple times.
+static bool current_filename_printed = false;
+
+/// True if we should print progress indicator and update it automatically
+/// if also verbose >= V_VERBOSE.
+static bool progress_automatic;
+
+/// True if message_progress_start() has been called but
+/// message_progress_end() hasn't been called yet.
+static bool progress_started = false;
+
+/// This is true when a progress message was printed and the cursor is still
+/// on the same line with the progress message. In that case, a newline has
+/// to be printed before any error messages.
+static bool progress_active = false;
+
+/// Pointer to lzma_stream used to do the encoding or decoding.
+static lzma_stream *progress_strm;
+
+/// This is true if we are in passthru mode (not actually compressing or
+/// decompressing) and thus cannot use lzma_get_progress(progress_strm, ...).
+/// That is, we are using coder_passthru() in coder.c.
+static bool progress_is_from_passthru;
+
+/// Expected size of the input stream is needed to show completion percentage
+/// and estimate remaining time.
+static uint64_t expected_in_size;
+
+
+// Use alarm() and SIGALRM when they are supported. This has two minor
+// advantages over the alternative of polling gettimeofday():
+// - It is possible for the user to send SIGINFO, SIGUSR1, or SIGALRM to
+// get intermediate progress information even when --verbose wasn't used
+// or stderr is not a terminal.
+// - alarm() + SIGALRM seems to have slightly less overhead than polling
+// gettimeofday().
+#ifdef SIGALRM
+
+const int message_progress_sigs[] = {
+ SIGALRM,
+#ifdef SIGINFO
+ SIGINFO,
+#endif
+#ifdef SIGUSR1
+ SIGUSR1,
+#endif
+ 0
+};
+
+/// The signal handler for SIGALRM sets this to true. It is set back to false
+/// once the progress message has been updated.
+static volatile sig_atomic_t progress_needs_updating = false;
+
+/// Signal handler for SIGALRM
+static void
+progress_signal_handler(int sig lzma_attribute((__unused__)))
+{
+ progress_needs_updating = true;
+ return;
+}
+
+#else
+
+/// This is true when progress message printing is wanted. Using the same
+/// variable name as above to avoid some ifdefs.
+static bool progress_needs_updating = false;
+
+/// Elapsed time when the next progress message update should be done.
+static uint64_t progress_next_update;
+
+#endif
+
+
+extern void
+message_init(void)
+{
+ // If --verbose is used, we use a progress indicator if and only
+ // if stderr is a terminal. If stderr is not a terminal, we print
+ // verbose information only after finishing the file. As a special
+ // exception, even if --verbose was not used, user can send SIGALRM
+ // to make us print progress information once without automatic
+ // updating.
+ progress_automatic = isatty(STDERR_FILENO);
+
+ // Commented out because COLUMNS is rarely exported to environment.
+ // Most users have at least 80 columns anyway, let's think something
+ // fancy here if enough people complain.
+/*
+ if (progress_automatic) {
+ // stderr is a terminal. Check the COLUMNS environment
+ // variable to see if the terminal is wide enough. If COLUMNS
+ // doesn't exist or it has some unparsable value, we assume
+ // that the terminal is wide enough.
+ const char *columns_str = getenv("COLUMNS");
+ if (columns_str != NULL) {
+ char *endptr;
+ const long columns = strtol(columns_str, &endptr, 10);
+ if (*endptr != '\0' || columns < 80)
+ progress_automatic = false;
+ }
+ }
+*/
+
+#ifdef SIGALRM
+ // Establish the signal handlers which set a flag to tell us that
+ // progress info should be updated.
+ struct sigaction sa;
+ sigemptyset(&sa.sa_mask);
+ sa.sa_flags = 0;
+ sa.sa_handler = &progress_signal_handler;
+
+ for (size_t i = 0; message_progress_sigs[i] != 0; ++i)
+ if (sigaction(message_progress_sigs[i], &sa, NULL))
+ message_signal_handler();
+#endif
+
+ return;
+}
+
+
+extern void
+message_verbosity_increase(void)
+{
+ if (verbosity < V_DEBUG)
+ ++verbosity;
+
+ return;
+}
+
+
+extern void
+message_verbosity_decrease(void)
+{
+ if (verbosity > V_SILENT)
+ --verbosity;
+
+ return;
+}
+
+
+extern enum message_verbosity
+message_verbosity_get(void)
+{
+ return verbosity;
+}
+
+
+extern void
+message_set_files(unsigned int files)
+{
+ files_total = files;
+ return;
+}
+
+
+/// Prints the name of the current file if it hasn't been printed already,
+/// except if we are processing exactly one stream from stdin to stdout.
+/// I think it looks nicer to not print "(stdin)" when --verbose is used
+/// in a pipe and no other files are processed.
+static void
+print_filename(void)
+{
+ if (!opt_robot && (files_total != 1 || filename != stdin_filename)) {
+ signals_block();
+
+ FILE *file = opt_mode == MODE_LIST ? stdout : stderr;
+
+ // If a file was already processed, put an empty line
+ // before the next filename to improve readability.
+ if (first_filename_printed)
+ fputc('\n', file);
+
+ first_filename_printed = true;
+ current_filename_printed = true;
+
+ // If we don't know how many files there will be due
+ // to usage of --files or --files0.
+ if (files_total == 0)
+ fprintf(file, "%s (%u)\n", filename,
+ files_pos);
+ else
+ fprintf(file, "%s (%u/%u)\n", filename,
+ files_pos, files_total);
+
+ signals_unblock();
+ }
+
+ return;
+}
+
+
+extern void
+message_filename(const char *src_name)
+{
+ // Start numbering the files starting from one.
+ ++files_pos;
+ filename = src_name;
+
+ if (verbosity >= V_VERBOSE
+ && (progress_automatic || opt_mode == MODE_LIST))
+ print_filename();
+ else
+ current_filename_printed = false;
+
+ return;
+}
+
+
+extern void
+message_progress_start(lzma_stream *strm, bool is_passthru, uint64_t in_size)
+{
+ // Store the pointer to the lzma_stream used to do the coding.
+ // It is needed to find out the position in the stream.
+ progress_strm = strm;
+ progress_is_from_passthru = is_passthru;
+
+ // Store the expected size of the file. If we aren't printing any
+ // statistics, then is will be unused. But since it is possible
+ // that the user sends us a signal to show statistics, we need
+ // to have it available anyway.
+ expected_in_size = in_size;
+
+ // Indicate that progress info may need to be printed before
+ // printing error messages.
+ progress_started = true;
+
+ // If progress indicator is wanted, print the filename and possibly
+ // the file count now.
+ if (verbosity >= V_VERBOSE && progress_automatic) {
+ // Start the timer to display the first progress message
+ // after one second. An alternative would be to show the
+ // first message almost immediately, but delaying by one
+ // second looks better to me, since extremely early
+ // progress info is pretty much useless.
+#ifdef SIGALRM
+ // First disable a possibly existing alarm.
+ alarm(0);
+ progress_needs_updating = false;
+ alarm(1);
+#else
+ progress_needs_updating = true;
+ progress_next_update = 1000;
+#endif
+ }
+
+ return;
+}
+
+
+/// Make the string indicating completion percentage.
+static const char *
+progress_percentage(uint64_t in_pos)
+{
+ // If the size of the input file is unknown or the size told us is
+ // clearly wrong since we have processed more data than the alleged
+ // size of the file, show a static string indicating that we have
+ // no idea of the completion percentage.
+ if (expected_in_size == 0 || in_pos > expected_in_size)
+ return "--- %";
+
+ // Never show 100.0 % before we actually are finished.
+ double percentage = (double)(in_pos) / (double)(expected_in_size)
+ * 99.9;
+
+ // Use big enough buffer to hold e.g. a multibyte decimal point.
+ static char buf[16];
+ snprintf(buf, sizeof(buf), "%.1f %%", percentage);
+
+ return buf;
+}
+
+
+/// Make the string containing the amount of input processed, amount of
+/// output produced, and the compression ratio.
+static const char *
+progress_sizes(uint64_t compressed_pos, uint64_t uncompressed_pos, bool final)
+{
+ // Use big enough buffer to hold e.g. a multibyte thousand separators.
+ static char buf[128];
+ char *pos = buf;
+ size_t left = sizeof(buf);
+
+ // Print the sizes. If this the final message, use more reasonable
+ // units than MiB if the file was small.
+ const enum nicestr_unit unit_min = final ? NICESTR_B : NICESTR_MIB;
+ my_snprintf(&pos, &left, "%s / %s",
+ uint64_to_nicestr(compressed_pos,
+ unit_min, NICESTR_TIB, false, 0),
+ uint64_to_nicestr(uncompressed_pos,
+ unit_min, NICESTR_TIB, false, 1));
+
+ // Avoid division by zero. If we cannot calculate the ratio, set
+ // it to some nice number greater than 10.0 so that it gets caught
+ // in the next if-clause.
+ const double ratio = uncompressed_pos > 0
+ ? (double)(compressed_pos) / (double)(uncompressed_pos)
+ : 16.0;
+
+ // If the ratio is very bad, just indicate that it is greater than
+ // 9.999. This way the length of the ratio field stays fixed.
+ if (ratio > 9.999)
+ snprintf(pos, left, " > %.3f", 9.999);
+ else
+ snprintf(pos, left, " = %.3f", ratio);
+
+ return buf;
+}
+
+
+/// Make the string containing the processing speed of uncompressed data.
+static const char *
+progress_speed(uint64_t uncompressed_pos, uint64_t elapsed)
+{
+ // Don't print the speed immediately, since the early values look
+ // somewhat random.
+ if (elapsed < 3000)
+ return "";
+
+ // The first character of KiB/s, MiB/s, or GiB/s:
+ static const char unit[] = { 'K', 'M', 'G' };
+
+ size_t unit_index = 0;
+
+ // Calculate the speed as KiB/s.
+ double speed = (double)(uncompressed_pos)
+ / ((double)(elapsed) * (1024.0 / 1000.0));
+
+ // Adjust the unit of the speed if needed.
+ while (speed > 999.0) {
+ speed /= 1024.0;
+ if (++unit_index == ARRAY_SIZE(unit))
+ return ""; // Way too fast ;-)
+ }
+
+ // Use decimal point only if the number is small. Examples:
+ // - 0.1 KiB/s
+ // - 9.9 KiB/s
+ // - 99 KiB/s
+ // - 999 KiB/s
+ // Use big enough buffer to hold e.g. a multibyte decimal point.
+ static char buf[16];
+ snprintf(buf, sizeof(buf), "%.*f %ciB/s",
+ speed > 9.9 ? 0 : 1, speed, unit[unit_index]);
+ return buf;
+}
+
+
+/// Make a string indicating elapsed time. The format is either
+/// M:SS or H:MM:SS depending on if the time is an hour or more.
+static const char *
+progress_time(uint64_t mseconds)
+{
+ // 9999 hours = 416 days
+ static char buf[sizeof("9999:59:59")];
+
+ // 32-bit variable is enough for elapsed time (136 years).
+ uint32_t seconds = (uint32_t)(mseconds / 1000);
+
+ // Don't show anything if the time is zero or ridiculously big.
+ if (seconds == 0 || seconds > ((9999 * 60) + 59) * 60 + 59)
+ return "";
+
+ uint32_t minutes = seconds / 60;
+ seconds %= 60;
+
+ if (minutes >= 60) {
+ const uint32_t hours = minutes / 60;
+ minutes %= 60;
+ snprintf(buf, sizeof(buf),
+ "%" PRIu32 ":%02" PRIu32 ":%02" PRIu32,
+ hours, minutes, seconds);
+ } else {
+ snprintf(buf, sizeof(buf), "%" PRIu32 ":%02" PRIu32,
+ minutes, seconds);
+ }
+
+ return buf;
+}
+
+
+/// Return a string containing estimated remaining time when
+/// reasonably possible.
+static const char *
+progress_remaining(uint64_t in_pos, uint64_t elapsed)
+{
+ // Don't show the estimated remaining time when it wouldn't
+ // make sense:
+ // - Input size is unknown.
+ // - Input has grown bigger since we started (de)compressing.
+ // - We haven't processed much data yet, so estimate would be
+ // too inaccurate.
+ // - Only a few seconds has passed since we started (de)compressing,
+ // so estimate would be too inaccurate.
+ if (expected_in_size == 0 || in_pos > expected_in_size
+ || in_pos < (UINT64_C(1) << 19) || elapsed < 8000)
+ return "";
+
+ // Calculate the estimate. Don't give an estimate of zero seconds,
+ // since it is possible that all the input has been already passed
+ // to the library, but there is still quite a bit of output pending.
+ uint32_t remaining = (uint32_t)((double)(expected_in_size - in_pos)
+ * ((double)(elapsed) / 1000.0) / (double)(in_pos));
+ if (remaining < 1)
+ remaining = 1;
+
+ static char buf[sizeof("9 h 55 min")];
+
+ // Select appropriate precision for the estimated remaining time.
+ if (remaining <= 10) {
+ // A maximum of 10 seconds remaining.
+ // Show the number of seconds as is.
+ snprintf(buf, sizeof(buf), "%" PRIu32 " s", remaining);
+
+ } else if (remaining <= 50) {
+ // A maximum of 50 seconds remaining.
+ // Round up to the next multiple of five seconds.
+ remaining = (remaining + 4) / 5 * 5;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " s", remaining);
+
+ } else if (remaining <= 590) {
+ // A maximum of 9 minutes and 50 seconds remaining.
+ // Round up to the next multiple of ten seconds.
+ remaining = (remaining + 9) / 10 * 10;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " min %" PRIu32 " s",
+ remaining / 60, remaining % 60);
+
+ } else if (remaining <= 59 * 60) {
+ // A maximum of 59 minutes remaining.
+ // Round up to the next multiple of a minute.
+ remaining = (remaining + 59) / 60;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " min", remaining);
+
+ } else if (remaining <= 9 * 3600 + 50 * 60) {
+ // A maximum of 9 hours and 50 minutes left.
+ // Round up to the next multiple of ten minutes.
+ remaining = (remaining + 599) / 600 * 10;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " h %" PRIu32 " min",
+ remaining / 60, remaining % 60);
+
+ } else if (remaining <= 23 * 3600) {
+ // A maximum of 23 hours remaining.
+ // Round up to the next multiple of an hour.
+ remaining = (remaining + 3599) / 3600;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " h", remaining);
+
+ } else if (remaining <= 9 * 24 * 3600 + 23 * 3600) {
+ // A maximum of 9 days and 23 hours remaining.
+ // Round up to the next multiple of an hour.
+ remaining = (remaining + 3599) / 3600;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " d %" PRIu32 " h",
+ remaining / 24, remaining % 24);
+
+ } else if (remaining <= 999 * 24 * 3600) {
+ // A maximum of 999 days remaining. ;-)
+ // Round up to the next multiple of a day.
+ remaining = (remaining + 24 * 3600 - 1) / (24 * 3600);
+ snprintf(buf, sizeof(buf), "%" PRIu32 " d", remaining);
+
+ } else {
+ // The estimated remaining time is too big. Don't show it.
+ return "";
+ }
+
+ return buf;
+}
+
+
+/// Get how much uncompressed and compressed data has been processed.
+static void
+progress_pos(uint64_t *in_pos,
+ uint64_t *compressed_pos, uint64_t *uncompressed_pos)
+{
+ uint64_t out_pos;
+ if (progress_is_from_passthru) {
+ // In passthru mode the progress info is in total_in/out but
+ // the *progress_strm itself isn't initialized and thus we
+ // cannot use lzma_get_progress().
+ *in_pos = progress_strm->total_in;
+ out_pos = progress_strm->total_out;
+ } else {
+ lzma_get_progress(progress_strm, in_pos, &out_pos);
+ }
+
+ // It cannot have processed more input than it has been given.
+ assert(*in_pos <= progress_strm->total_in);
+
+ // It cannot have produced more output than it claims to have ready.
+ assert(out_pos >= progress_strm->total_out);
+
+ if (opt_mode == MODE_COMPRESS) {
+ *compressed_pos = out_pos;
+ *uncompressed_pos = *in_pos;
+ } else {
+ *compressed_pos = *in_pos;
+ *uncompressed_pos = out_pos;
+ }
+
+ return;
+}
+
+
+extern void
+message_progress_update(void)
+{
+ if (!progress_needs_updating)
+ return;
+
+ // Calculate how long we have been processing this file.
+ const uint64_t elapsed = mytime_get_elapsed();
+
+#ifndef SIGALRM
+ if (progress_next_update > elapsed)
+ return;
+
+ progress_next_update = elapsed + 1000;
+#endif
+
+ // Get our current position in the stream.
+ uint64_t in_pos;
+ uint64_t compressed_pos;
+ uint64_t uncompressed_pos;
+ progress_pos(&in_pos, &compressed_pos, &uncompressed_pos);
+
+ // Block signals so that fprintf() doesn't get interrupted.
+ signals_block();
+
+ // Print the filename if it hasn't been printed yet.
+ if (!current_filename_printed)
+ print_filename();
+
+ // Print the actual progress message. The idea is that there is at
+ // least three spaces between the fields in typical situations, but
+ // even in rare situations there is at least one space.
+ const char *cols[5] = {
+ progress_percentage(in_pos),
+ progress_sizes(compressed_pos, uncompressed_pos, false),
+ progress_speed(uncompressed_pos, elapsed),
+ progress_time(elapsed),
+ progress_remaining(in_pos, elapsed),
+ };
+ fprintf(stderr, "\r %*s %*s %*s %10s %10s\r",
+ tuklib_mbstr_fw(cols[0], 6), cols[0],
+ tuklib_mbstr_fw(cols[1], 35), cols[1],
+ tuklib_mbstr_fw(cols[2], 9), cols[2],
+ cols[3],
+ cols[4]);
+
+#ifdef SIGALRM
+ // Updating the progress info was finished. Reset
+ // progress_needs_updating to wait for the next SIGALRM.
+ //
+ // NOTE: This has to be done before alarm(1) or with (very) bad
+ // luck we could be setting this to false after the alarm has already
+ // been triggered.
+ progress_needs_updating = false;
+
+ if (verbosity >= V_VERBOSE && progress_automatic) {
+ // Mark that the progress indicator is active, so if an error
+ // occurs, the error message gets printed cleanly.
+ progress_active = true;
+
+ // Restart the timer so that progress_needs_updating gets
+ // set to true after about one second.
+ alarm(1);
+ } else {
+ // The progress message was printed because user had sent us
+ // SIGALRM. In this case, each progress message is printed
+ // on its own line.
+ fputc('\n', stderr);
+ }
+#else
+ // When SIGALRM isn't supported and we get here, it's always due to
+ // automatic progress update. We set progress_active here too like
+ // described above.
+ assert(verbosity >= V_VERBOSE);
+ assert(progress_automatic);
+ progress_active = true;
+#endif
+
+ signals_unblock();
+
+ return;
+}
+
+
+static void
+progress_flush(bool finished)
+{
+ if (!progress_started || verbosity < V_VERBOSE)
+ return;
+
+ uint64_t in_pos;
+ uint64_t compressed_pos;
+ uint64_t uncompressed_pos;
+ progress_pos(&in_pos, &compressed_pos, &uncompressed_pos);
+
+ // Avoid printing intermediate progress info if some error occurs
+ // in the beginning of the stream. (If something goes wrong later in
+ // the stream, it is sometimes useful to tell the user where the
+ // error approximately occurred, especially if the error occurs
+ // after a time-consuming operation.)
+ if (!finished && !progress_active
+ && (compressed_pos == 0 || uncompressed_pos == 0))
+ return;
+
+ progress_active = false;
+
+ const uint64_t elapsed = mytime_get_elapsed();
+
+ signals_block();
+
+ // When using the auto-updating progress indicator, the final
+ // statistics are printed in the same format as the progress
+ // indicator itself.
+ if (progress_automatic) {
+ const char *cols[5] = {
+ finished ? "100 %" : progress_percentage(in_pos),
+ progress_sizes(compressed_pos, uncompressed_pos, true),
+ progress_speed(uncompressed_pos, elapsed),
+ progress_time(elapsed),
+ finished ? "" : progress_remaining(in_pos, elapsed),
+ };
+ fprintf(stderr, "\r %*s %*s %*s %10s %10s\n",
+ tuklib_mbstr_fw(cols[0], 6), cols[0],
+ tuklib_mbstr_fw(cols[1], 35), cols[1],
+ tuklib_mbstr_fw(cols[2], 9), cols[2],
+ cols[3],
+ cols[4]);
+ } else {
+ // The filename is always printed.
+ fprintf(stderr, _("%s: "), filename);
+
+ // Percentage is printed only if we didn't finish yet.
+ if (!finished) {
+ // Don't print the percentage when it isn't known
+ // (starts with a dash).
+ const char *percentage = progress_percentage(in_pos);
+ if (percentage[0] != '-')
+ fprintf(stderr, "%s, ", percentage);
+ }
+
+ // Size information is always printed.
+ fprintf(stderr, "%s", progress_sizes(
+ compressed_pos, uncompressed_pos, true));
+
+ // The speed and elapsed time aren't always shown.
+ const char *speed = progress_speed(uncompressed_pos, elapsed);
+ if (speed[0] != '\0')
+ fprintf(stderr, ", %s", speed);
+
+ const char *elapsed_str = progress_time(elapsed);
+ if (elapsed_str[0] != '\0')
+ fprintf(stderr, ", %s", elapsed_str);
+
+ fputc('\n', stderr);
+ }
+
+ signals_unblock();
+
+ return;
+}
+
+
+extern void
+message_progress_end(bool success)
+{
+ assert(progress_started);
+ progress_flush(success);
+ progress_started = false;
+ return;
+}
+
+
+static void
+vmessage(enum message_verbosity v, const char *fmt, va_list ap)
+{
+ if (v <= verbosity) {
+ signals_block();
+
+ progress_flush(false);
+
+ // TRANSLATORS: This is the program name in the beginning
+ // of the line in messages. Usually it becomes "xz: ".
+ // This is a translatable string because French needs
+ // a space before a colon.
+ fprintf(stderr, _("%s: "), progname);
+
+#ifdef __clang__
+# pragma GCC diagnostic push
+# pragma GCC diagnostic ignored "-Wformat-nonliteral"
+#endif
+ vfprintf(stderr, fmt, ap);
+#ifdef __clang__
+# pragma GCC diagnostic pop
+#endif
+
+ fputc('\n', stderr);
+
+ signals_unblock();
+ }
+
+ return;
+}
+
+
+extern void
+message(enum message_verbosity v, const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ vmessage(v, fmt, ap);
+ va_end(ap);
+ return;
+}
+
+
+extern void
+message_warning(const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ vmessage(V_WARNING, fmt, ap);
+ va_end(ap);
+
+ set_exit_status(E_WARNING);
+ return;
+}
+
+
+extern void
+message_error(const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ vmessage(V_ERROR, fmt, ap);
+ va_end(ap);
+
+ set_exit_status(E_ERROR);
+ return;
+}
+
+
+extern void
+message_fatal(const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ vmessage(V_ERROR, fmt, ap);
+ va_end(ap);
+
+ tuklib_exit(E_ERROR, E_ERROR, false);
+}
+
+
+extern void
+message_bug(void)
+{
+ message_fatal(_("Internal error (bug)"));
+}
+
+
+extern void
+message_signal_handler(void)
+{
+ message_fatal(_("Cannot establish signal handlers"));
+}
+
+
+extern const char *
+message_strm(lzma_ret code)
+{
+ switch (code) {
+ case LZMA_NO_CHECK:
+ return _("No integrity check; not verifying file integrity");
+
+ case LZMA_UNSUPPORTED_CHECK:
+ return _("Unsupported type of integrity check; "
+ "not verifying file integrity");
+
+ case LZMA_MEM_ERROR:
+ return strerror(ENOMEM);
+
+ case LZMA_MEMLIMIT_ERROR:
+ return _("Memory usage limit reached");
+
+ case LZMA_FORMAT_ERROR:
+ return _("File format not recognized");
+
+ case LZMA_OPTIONS_ERROR:
+ return _("Unsupported options");
+
+ case LZMA_DATA_ERROR:
+ return _("Compressed data is corrupt");
+
+ case LZMA_BUF_ERROR:
+ return _("Unexpected end of input");
+
+ case LZMA_OK:
+ case LZMA_STREAM_END:
+ case LZMA_GET_CHECK:
+ case LZMA_PROG_ERROR:
+ case LZMA_SEEK_NEEDED:
+ case LZMA_RET_INTERNAL1:
+ case LZMA_RET_INTERNAL2:
+ case LZMA_RET_INTERNAL3:
+ case LZMA_RET_INTERNAL4:
+ case LZMA_RET_INTERNAL5:
+ case LZMA_RET_INTERNAL6:
+ case LZMA_RET_INTERNAL7:
+ case LZMA_RET_INTERNAL8:
+ // Without "default", compiler will warn if new constants
+ // are added to lzma_ret, it is not too easy to forget to
+ // add the new constants to this function.
+ break;
+ }
+
+ return _("Internal error (bug)");
+}
+
+
+extern void
+message_mem_needed(enum message_verbosity v, uint64_t memusage)
+{
+ if (v > verbosity)
+ return;
+
+ // Convert memusage to MiB, rounding up to the next full MiB.
+ // This way the user can always use the displayed usage as
+ // the new memory usage limit. (If we rounded to the nearest,
+ // the user might need to +1 MiB to get high enough limit.)
+ memusage = round_up_to_mib(memusage);
+
+ uint64_t memlimit = hardware_memlimit_get(opt_mode);
+
+ // Handle the case when there is no memory usage limit.
+ // This way we don't print a weird message with a huge number.
+ if (memlimit == UINT64_MAX) {
+ message(v, _("%s MiB of memory is required. "
+ "The limiter is disabled."),
+ uint64_to_str(memusage, 0));
+ return;
+ }
+
+ // With US-ASCII:
+ // 2^64 with thousand separators + " MiB" suffix + '\0' = 26 + 4 + 1
+ // But there may be multibyte chars so reserve enough space.
+ char memlimitstr[128];
+
+ // Show the memory usage limit as MiB unless it is less than 1 MiB.
+ // This way it's easy to notice errors where one has typed
+ // --memory=123 instead of --memory=123MiB.
+ if (memlimit < (UINT32_C(1) << 20)) {
+ snprintf(memlimitstr, sizeof(memlimitstr), "%s B",
+ uint64_to_str(memlimit, 1));
+ } else {
+ // Round up just like with memusage. If this function is
+ // called for informational purposes (to just show the
+ // current usage and limit), we should never show that
+ // the usage is higher than the limit, which would give
+ // a false impression that the memory usage limit isn't
+ // properly enforced.
+ snprintf(memlimitstr, sizeof(memlimitstr), "%s MiB",
+ uint64_to_str(round_up_to_mib(memlimit), 1));
+ }
+
+ message(v, _("%s MiB of memory is required. The limit is %s."),
+ uint64_to_str(memusage, 0), memlimitstr);
+
+ return;
+}
+
+
+extern void
+message_filters_show(enum message_verbosity v, const lzma_filter *filters)
+{
+ if (v > verbosity)
+ return;
+
+ char *buf;
+ const lzma_ret ret = lzma_str_from_filters(&buf, filters,
+ LZMA_STR_ENCODER | LZMA_STR_GETOPT_LONG, NULL);
+ if (ret != LZMA_OK)
+ message_fatal("%s", message_strm(ret));
+
+ fprintf(stderr, _("%s: Filter chain: %s\n"), progname, buf);
+ free(buf);
+ return;
+}
+
+
+extern void
+message_try_help(void)
+{
+ // Print this with V_WARNING instead of V_ERROR to prevent it from
+ // showing up when --quiet has been specified.
+ message(V_WARNING, _("Try `%s --help' for more information."),
+ progname);
+ return;
+}
+
+
+extern void
+message_version(void)
+{
+ // It is possible that liblzma version is different than the command
+ // line tool version, so print both.
+ if (opt_robot) {
+ printf("XZ_VERSION=%" PRIu32 "\nLIBLZMA_VERSION=%" PRIu32 "\n",
+ LZMA_VERSION, lzma_version_number());
+ } else {
+ printf("xz (" PACKAGE_NAME ") " LZMA_VERSION_STRING "\n");
+ printf("liblzma %s\n", lzma_version_string());
+ }
+
+ tuklib_exit(E_SUCCESS, E_ERROR, verbosity != V_SILENT);
+}
+
+
+extern void
+message_help(bool long_help)
+{
+ printf(_("Usage: %s [OPTION]... [FILE]...\n"
+ "Compress or decompress FILEs in the .xz format.\n\n"),
+ progname);
+
+ // NOTE: The short help doesn't currently have options that
+ // take arguments.
+ if (long_help)
+ puts(_("Mandatory arguments to long options are mandatory "
+ "for short options too.\n"));
+
+ if (long_help)
+ puts(_(" Operation mode:\n"));
+
+ puts(_(
+" -z, --compress force compression\n"
+" -d, --decompress force decompression\n"
+" -t, --test test compressed file integrity\n"
+" -l, --list list information about .xz files"));
+
+ if (long_help)
+ puts(_("\n Operation modifiers:\n"));
+
+ puts(_(
+" -k, --keep keep (don't delete) input files\n"
+" -f, --force force overwrite of output file and (de)compress links\n"
+" -c, --stdout write to standard output and don't delete input files"));
+ // NOTE: --to-stdout isn't included above because it's not
+ // the recommended spelling. It was copied from gzip but other
+ // compressors with gzip-like syntax don't support it.
+
+ if (long_help) {
+ puts(_(
+" --single-stream decompress only the first stream, and silently\n"
+" ignore possible remaining input data"));
+ puts(_(
+" --no-sparse do not create sparse files when decompressing\n"
+" -S, --suffix=.SUF use the suffix `.SUF' on compressed files\n"
+" --files[=FILE] read filenames to process from FILE; if FILE is\n"
+" omitted, filenames are read from the standard input;\n"
+" filenames must be terminated with the newline character\n"
+" --files0[=FILE] like --files but use the null character as terminator"));
+ }
+
+ if (long_help) {
+ puts(_("\n Basic file format and compression options:\n"));
+ puts(_(
+" -F, --format=FMT file format to encode or decode; possible values are\n"
+" `auto' (default), `xz', `lzma', `lzip', and `raw'\n"
+" -C, --check=CHECK integrity check type: `none' (use with caution),\n"
+" `crc32', `crc64' (default), or `sha256'"));
+ puts(_(
+" --ignore-check don't verify the integrity check when decompressing"));
+ }
+
+ puts(_(
+" -0 ... -9 compression preset; default is 6; take compressor *and*\n"
+" decompressor memory usage into account before using 7-9!"));
+
+ puts(_(
+" -e, --extreme try to improve compression ratio by using more CPU time;\n"
+" does not affect decompressor memory requirements"));
+
+ puts(_(
+" -T, --threads=NUM use at most NUM threads; the default is 1; set to 0\n"
+" to use as many threads as there are processor cores"));
+
+ if (long_help) {
+ puts(_(
+" --block-size=SIZE\n"
+" start a new .xz block after every SIZE bytes of input;\n"
+" use this to set the block size for threaded compression"));
+ puts(_(
+" --block-list=SIZES\n"
+" start a new .xz block after the given comma-separated\n"
+" intervals of uncompressed data"));
+ puts(_(
+" --flush-timeout=TIMEOUT\n"
+" when compressing, if more than TIMEOUT milliseconds has\n"
+" passed since the previous flush and reading more input\n"
+" would block, all pending data is flushed out"
+ ));
+ puts(_( // xgettext:no-c-format
+" --memlimit-compress=LIMIT\n"
+" --memlimit-decompress=LIMIT\n"
+" --memlimit-mt-decompress=LIMIT\n"
+" -M, --memlimit=LIMIT\n"
+" set memory usage limit for compression, decompression,\n"
+" threaded decompression, or all of these; LIMIT is in\n"
+" bytes, % of RAM, or 0 for defaults"));
+
+ puts(_(
+" --no-adjust if compression settings exceed the memory usage limit,\n"
+" give an error instead of adjusting the settings downwards"));
+ }
+
+ if (long_help) {
+ puts(_(
+"\n Custom filter chain for compression (alternative for using presets):"));
+
+#if defined(HAVE_ENCODER_LZMA1) || defined(HAVE_DECODER_LZMA1) \
+ || defined(HAVE_ENCODER_LZMA2) || defined(HAVE_DECODER_LZMA2)
+ // TRANSLATORS: The word "literal" in "literal context bits"
+ // means how many "context bits" to use when encoding
+ // literals. A literal is a single 8-bit byte. It doesn't
+ // mean "literally" here.
+ puts(_(
+"\n"
+" --lzma1[=OPTS] LZMA1 or LZMA2; OPTS is a comma-separated list of zero or\n"
+" --lzma2[=OPTS] more of the following options (valid values; default):\n"
+" preset=PRE reset options to a preset (0-9[e])\n"
+" dict=NUM dictionary size (4KiB - 1536MiB; 8MiB)\n"
+" lc=NUM number of literal context bits (0-4; 3)\n"
+" lp=NUM number of literal position bits (0-4; 0)\n"
+" pb=NUM number of position bits (0-4; 2)\n"
+" mode=MODE compression mode (fast, normal; normal)\n"
+" nice=NUM nice length of a match (2-273; 64)\n"
+" mf=NAME match finder (hc3, hc4, bt2, bt3, bt4; bt4)\n"
+" depth=NUM maximum search depth; 0=automatic (default)"));
+#endif
+
+ puts(_(
+"\n"
+" --x86[=OPTS] x86 BCJ filter (32-bit and 64-bit)\n"
+" --arm[=OPTS] ARM BCJ filter\n"
+" --armthumb[=OPTS] ARM-Thumb BCJ filter\n"
+" --arm64[=OPTS] ARM64 BCJ filter\n"
+" --powerpc[=OPTS] PowerPC BCJ filter (big endian only)\n"
+" --ia64[=OPTS] IA-64 (Itanium) BCJ filter\n"
+" --sparc[=OPTS] SPARC BCJ filter\n"
+" Valid OPTS for all BCJ filters:\n"
+" start=NUM start offset for conversions (default=0)"));
+
+#if defined(HAVE_ENCODER_DELTA) || defined(HAVE_DECODER_DELTA)
+ puts(_(
+"\n"
+" --delta[=OPTS] Delta filter; valid OPTS (valid values; default):\n"
+" dist=NUM distance between bytes being subtracted\n"
+" from each other (1-256; 1)"));
+#endif
+ }
+
+ if (long_help)
+ puts(_("\n Other options:\n"));
+
+ puts(_(
+" -q, --quiet suppress warnings; specify twice to suppress errors too\n"
+" -v, --verbose be verbose; specify twice for even more verbose"));
+
+ if (long_help) {
+ puts(_(
+" -Q, --no-warn make warnings not affect the exit status"));
+ puts(_(
+" --robot use machine-parsable messages (useful for scripts)"));
+ puts("");
+ puts(_(
+" --info-memory display the total amount of RAM and the currently active\n"
+" memory usage limits, and exit"));
+ puts(_(
+" -h, --help display the short help (lists only the basic options)\n"
+" -H, --long-help display this long help and exit"));
+ } else {
+ puts(_(
+" -h, --help display this short help and exit\n"
+" -H, --long-help display the long help (lists also the advanced options)"));
+ }
+
+ puts(_(
+" -V, --version display the version number and exit"));
+
+ puts(_("\nWith no FILE, or when FILE is -, read standard input.\n"));
+
+ // TRANSLATORS: This message indicates the bug reporting address
+ // for this package. Please add _another line_ saying
+ // "Report translation bugs to <...>\n" with the email or WWW
+ // address for translation bugs. Thanks.
+ printf(_("Report bugs to <%s> (in English or Finnish).\n"),
+ PACKAGE_BUGREPORT);
+ printf(_("%s home page: <%s>\n"), PACKAGE_NAME, PACKAGE_URL);
+
+#if LZMA_VERSION_STABILITY != LZMA_VERSION_STABILITY_STABLE
+ puts(_(
+"THIS IS A DEVELOPMENT VERSION NOT INTENDED FOR PRODUCTION USE."));
+#endif
+
+ tuklib_exit(E_SUCCESS, E_ERROR, verbosity != V_SILENT);
+}
diff --git a/src/xz/message.h b/src/xz/message.h
new file mode 100644
index 0000000..f608ec7
--- /dev/null
+++ b/src/xz/message.h
@@ -0,0 +1,156 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file message.h
+/// \brief Printing messages to stderr
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// Verbosity levels
+enum message_verbosity {
+ V_SILENT, ///< No messages
+ V_ERROR, ///< Only error messages
+ V_WARNING, ///< Errors and warnings
+ V_VERBOSE, ///< Errors, warnings, and verbose statistics
+ V_DEBUG, ///< Very verbose
+};
+
+
+/// \brief Signals used for progress message handling
+extern const int message_progress_sigs[];
+
+
+/// \brief Initializes the message functions
+///
+/// If an error occurs, this function doesn't return.
+///
+extern void message_init(void);
+
+
+/// Increase verbosity level by one step unless it was at maximum.
+extern void message_verbosity_increase(void);
+
+/// Decrease verbosity level by one step unless it was at minimum.
+extern void message_verbosity_decrease(void);
+
+/// Get the current verbosity level.
+extern enum message_verbosity message_verbosity_get(void);
+
+
+/// \brief Print a message if verbosity level is at least "verbosity"
+///
+/// This doesn't touch the exit status.
+lzma_attribute((__format__(__printf__, 2, 3)))
+extern void message(enum message_verbosity verbosity, const char *fmt, ...);
+
+
+/// \brief Prints a warning and possibly sets exit status
+///
+/// The message is printed only if verbosity level is at least V_WARNING.
+/// The exit status is set to WARNING unless it was already at ERROR.
+lzma_attribute((__format__(__printf__, 1, 2)))
+extern void message_warning(const char *fmt, ...);
+
+
+/// \brief Prints an error message and sets exit status
+///
+/// The message is printed only if verbosity level is at least V_ERROR.
+/// The exit status is set to ERROR.
+lzma_attribute((__format__(__printf__, 1, 2)))
+extern void message_error(const char *fmt, ...);
+
+
+/// \brief Prints an error message and exits with EXIT_ERROR
+///
+/// The message is printed only if verbosity level is at least V_ERROR.
+tuklib_attr_noreturn
+lzma_attribute((__format__(__printf__, 1, 2)))
+extern void message_fatal(const char *fmt, ...);
+
+
+/// Print an error message that an internal error occurred and exit with
+/// EXIT_ERROR.
+tuklib_attr_noreturn
+extern void message_bug(void);
+
+
+/// Print a message that establishing signal handlers failed, and exit with
+/// exit status ERROR.
+tuklib_attr_noreturn
+extern void message_signal_handler(void);
+
+
+/// Convert lzma_ret to a string.
+extern const char *message_strm(lzma_ret code);
+
+
+/// Display how much memory was needed and how much the limit was.
+extern void message_mem_needed(enum message_verbosity v, uint64_t memusage);
+
+
+/// Print the filter chain.
+extern void message_filters_show(
+ enum message_verbosity v, const lzma_filter *filters);
+
+
+/// Print a message that user should try --help.
+extern void message_try_help(void);
+
+
+/// Prints the version number to stdout and exits with exit status SUCCESS.
+tuklib_attr_noreturn
+extern void message_version(void);
+
+
+/// Print the help message.
+tuklib_attr_noreturn
+extern void message_help(bool long_help);
+
+
+/// \brief Set the total number of files to be processed
+///
+/// Standard input is counted as a file here. This is used when printing
+/// the filename via message_filename().
+extern void message_set_files(unsigned int files);
+
+
+/// \brief Set the name of the current file and possibly print it too
+///
+/// The name is printed immediately if --list was used or if --verbose
+/// was used and stderr is a terminal. Even when the filename isn't printed,
+/// it is stored so that it can be printed later if needed for progress
+/// messages.
+extern void message_filename(const char *src_name);
+
+
+/// \brief Start progress info handling
+///
+/// message_filename() must be called before this function to set
+/// the filename.
+///
+/// This must be paired with a call to message_progress_end() before the
+/// given *strm becomes invalid.
+///
+/// \param strm Pointer to lzma_stream used for the coding.
+/// \param in_size Size of the input file, or zero if unknown.
+///
+extern void message_progress_start(lzma_stream *strm,
+ bool is_passthru, uint64_t in_size);
+
+
+/// Update the progress info if in verbose mode and enough time has passed
+/// since the previous update. This can be called only when
+/// message_progress_start() has already been used.
+extern void message_progress_update(void);
+
+
+/// \brief Finishes the progress message if we were in verbose mode
+///
+/// \param finished True if the whole stream was successfully coded
+/// and output written to the output stream.
+///
+extern void message_progress_end(bool finished);
diff --git a/src/xz/mytime.c b/src/xz/mytime.c
new file mode 100644
index 0000000..7e8a074
--- /dev/null
+++ b/src/xz/mytime.c
@@ -0,0 +1,86 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file mytime.c
+/// \brief Time handling functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+#if defined(HAVE_CLOCK_GETTIME) && defined(HAVE_CLOCK_MONOTONIC)
+# include <time.h>
+#else
+# include <sys/time.h>
+#endif
+
+uint64_t opt_flush_timeout = 0;
+
+static uint64_t start_time;
+static uint64_t next_flush;
+
+
+/// \brief Get the current time as milliseconds
+///
+/// It's relative to some point but not necessarily to the UNIX Epoch.
+static uint64_t
+mytime_now(void)
+{
+#if defined(HAVE_CLOCK_GETTIME) && defined(HAVE_CLOCK_MONOTONIC)
+ // If CLOCK_MONOTONIC was available at compile time but for some
+ // reason isn't at runtime, fallback to CLOCK_REALTIME which
+ // according to POSIX is mandatory for all implementations.
+ static clockid_t clk_id = CLOCK_MONOTONIC;
+ struct timespec tv;
+ while (clock_gettime(clk_id, &tv))
+ clk_id = CLOCK_REALTIME;
+
+ return (uint64_t)tv.tv_sec * 1000 + (uint64_t)(tv.tv_nsec / 1000000);
+#else
+ struct timeval tv;
+ gettimeofday(&tv, NULL);
+ return (uint64_t)tv.tv_sec * 1000 + (uint64_t)(tv.tv_usec / 1000);
+#endif
+}
+
+
+extern void
+mytime_set_start_time(void)
+{
+ start_time = mytime_now();
+ return;
+}
+
+
+extern uint64_t
+mytime_get_elapsed(void)
+{
+ return mytime_now() - start_time;
+}
+
+
+extern void
+mytime_set_flush_time(void)
+{
+ next_flush = mytime_now() + opt_flush_timeout;
+ return;
+}
+
+
+extern int
+mytime_get_flush_timeout(void)
+{
+ if (opt_flush_timeout == 0 || opt_mode != MODE_COMPRESS)
+ return -1;
+
+ const uint64_t now = mytime_now();
+ if (now >= next_flush)
+ return 0;
+
+ const uint64_t remaining = next_flush - now;
+ return remaining > INT_MAX ? INT_MAX : (int)remaining;
+}
diff --git a/src/xz/mytime.h b/src/xz/mytime.h
new file mode 100644
index 0000000..a7be2aa
--- /dev/null
+++ b/src/xz/mytime.h
@@ -0,0 +1,43 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file mytime.h
+/// \brief Time handling functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+
+/// \brief Number of milliseconds to between LZMA_SYNC_FLUSHes
+///
+/// If 0, timed flushing is disabled. Otherwise if no more input is available
+/// and not at the end of the file and at least opt_flush_timeout milliseconds
+/// has elapsed since the start of compression or the previous flushing
+/// (LZMA_SYNC_FLUSH or LZMA_FULL_FLUSH), set LZMA_SYNC_FLUSH to flush
+/// the pending data.
+extern uint64_t opt_flush_timeout;
+
+
+/// \brief Store the time when (de)compression was started
+///
+/// The start time is also stored as the time of the first flush.
+extern void mytime_set_start_time(void);
+
+
+/// \brief Get the number of milliseconds since the operation started
+extern uint64_t mytime_get_elapsed(void);
+
+
+/// \brief Store the time of when compressor was flushed
+extern void mytime_set_flush_time(void);
+
+
+/// \brief Get the number of milliseconds until the next flush
+///
+/// This returns -1 if no timed flushing is used.
+///
+/// The return value is intended for use with poll().
+extern int mytime_get_flush_timeout(void);
diff --git a/src/xz/options.c b/src/xz/options.c
new file mode 100644
index 0000000..4d5e899
--- /dev/null
+++ b/src/xz/options.c
@@ -0,0 +1,359 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file options.c
+/// \brief Parser for filter-specific options
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+
+///////////////////
+// Generic stuff //
+///////////////////
+
+typedef struct {
+ const char *name;
+ uint64_t id;
+} name_id_map;
+
+
+typedef struct {
+ const char *name;
+ const name_id_map *map;
+ uint64_t min;
+ uint64_t max;
+} option_map;
+
+
+/// Parses option=value pairs that are separated with commas:
+/// opt=val,opt=val,opt=val
+///
+/// Each option is a string, that is converted to an integer using the
+/// index where the option string is in the array.
+///
+/// Value can be
+/// - a string-id map mapping a list of possible string values to integers
+/// (opts[i].map != NULL, opts[i].min and opts[i].max are ignored);
+/// - a number with minimum and maximum value limit
+/// (opts[i].map == NULL && opts[i].min != UINT64_MAX);
+/// - a string that will be parsed by the filter-specific code
+/// (opts[i].map == NULL && opts[i].min == UINT64_MAX, opts[i].max ignored)
+///
+/// When parsing both option and value succeed, a filter-specific function
+/// is called, which should update the given value to filter-specific
+/// options structure.
+///
+/// This returns only if no errors occur.
+///
+/// \param str String containing the options from the command line
+/// \param opts Filter-specific option map
+/// \param set Filter-specific function to update filter_options
+/// \param filter_options Pointer to filter-specific options structure
+///
+static void
+parse_options(const char *str, const option_map *opts,
+ void (*set)(void *filter_options,
+ unsigned key, uint64_t value, const char *valuestr),
+ void *filter_options)
+{
+ if (str == NULL || str[0] == '\0')
+ return;
+
+ char *s = xstrdup(str);
+ char *name = s;
+
+ while (*name != '\0') {
+ if (*name == ',') {
+ ++name;
+ continue;
+ }
+
+ char *split = strchr(name, ',');
+ if (split != NULL)
+ *split = '\0';
+
+ char *value = strchr(name, '=');
+ if (value != NULL)
+ *value++ = '\0';
+
+ if (value == NULL || value[0] == '\0')
+ message_fatal(_("%s: Options must be `name=value' "
+ "pairs separated with commas"), str);
+
+ // Look for the option name from the option map.
+ unsigned i = 0;
+ while (true) {
+ if (opts[i].name == NULL)
+ message_fatal(_("%s: Invalid option name"),
+ name);
+
+ if (strcmp(name, opts[i].name) == 0)
+ break;
+
+ ++i;
+ }
+
+ // Option was found from the map. See how we should handle it.
+ if (opts[i].map != NULL) {
+ // value is a string which we should map
+ // to an integer.
+ unsigned j;
+ for (j = 0; opts[i].map[j].name != NULL; ++j) {
+ if (strcmp(opts[i].map[j].name, value) == 0)
+ break;
+ }
+
+ if (opts[i].map[j].name == NULL)
+ message_fatal(_("%s: Invalid option value"),
+ value);
+
+ set(filter_options, i, opts[i].map[j].id, value);
+
+ } else if (opts[i].min == UINT64_MAX) {
+ // value is a special string that will be
+ // parsed by set().
+ set(filter_options, i, 0, value);
+
+ } else {
+ // value is an integer.
+ const uint64_t v = str_to_uint64(name, value,
+ opts[i].min, opts[i].max);
+ set(filter_options, i, v, value);
+ }
+
+ // Check if it was the last option.
+ if (split == NULL)
+ break;
+
+ name = split + 1;
+ }
+
+ free(s);
+ return;
+}
+
+
+///////////
+// Delta //
+///////////
+
+enum {
+ OPT_DIST,
+};
+
+
+static void
+set_delta(void *options, unsigned key, uint64_t value,
+ const char *valuestr lzma_attribute((__unused__)))
+{
+ lzma_options_delta *opt = options;
+ switch (key) {
+ case OPT_DIST:
+ opt->dist = value;
+ break;
+ }
+}
+
+
+extern lzma_options_delta *
+options_delta(const char *str)
+{
+ static const option_map opts[] = {
+ { "dist", NULL, LZMA_DELTA_DIST_MIN,
+ LZMA_DELTA_DIST_MAX },
+ { NULL, NULL, 0, 0 }
+ };
+
+ lzma_options_delta *options = xmalloc(sizeof(lzma_options_delta));
+ *options = (lzma_options_delta){
+ // It's hard to give a useful default for this.
+ .type = LZMA_DELTA_TYPE_BYTE,
+ .dist = LZMA_DELTA_DIST_MIN,
+ };
+
+ parse_options(str, opts, &set_delta, options);
+
+ return options;
+}
+
+
+/////////
+// BCJ //
+/////////
+
+enum {
+ OPT_START_OFFSET,
+};
+
+
+static void
+set_bcj(void *options, unsigned key, uint64_t value,
+ const char *valuestr lzma_attribute((__unused__)))
+{
+ lzma_options_bcj *opt = options;
+ switch (key) {
+ case OPT_START_OFFSET:
+ opt->start_offset = value;
+ break;
+ }
+}
+
+
+extern lzma_options_bcj *
+options_bcj(const char *str)
+{
+ static const option_map opts[] = {
+ { "start", NULL, 0, UINT32_MAX },
+ { NULL, NULL, 0, 0 }
+ };
+
+ lzma_options_bcj *options = xmalloc(sizeof(lzma_options_bcj));
+ *options = (lzma_options_bcj){
+ .start_offset = 0,
+ };
+
+ parse_options(str, opts, &set_bcj, options);
+
+ return options;
+}
+
+
+//////////
+// LZMA //
+//////////
+
+enum {
+ OPT_PRESET,
+ OPT_DICT,
+ OPT_LC,
+ OPT_LP,
+ OPT_PB,
+ OPT_MODE,
+ OPT_NICE,
+ OPT_MF,
+ OPT_DEPTH,
+};
+
+
+tuklib_attr_noreturn
+static void
+error_lzma_preset(const char *valuestr)
+{
+ message_fatal(_("Unsupported LZMA1/LZMA2 preset: %s"), valuestr);
+}
+
+
+static void
+set_lzma(void *options, unsigned key, uint64_t value, const char *valuestr)
+{
+ lzma_options_lzma *opt = options;
+
+ switch (key) {
+ case OPT_PRESET: {
+ if (valuestr[0] < '0' || valuestr[0] > '9')
+ error_lzma_preset(valuestr);
+
+ uint32_t preset = (uint32_t)(valuestr[0] - '0');
+
+ // Currently only "e" is supported as a modifier,
+ // so keep this simple for now.
+ if (valuestr[1] != '\0') {
+ if (valuestr[1] == 'e')
+ preset |= LZMA_PRESET_EXTREME;
+ else
+ error_lzma_preset(valuestr);
+
+ if (valuestr[2] != '\0')
+ error_lzma_preset(valuestr);
+ }
+
+ if (lzma_lzma_preset(options, preset))
+ error_lzma_preset(valuestr);
+
+ break;
+ }
+
+ case OPT_DICT:
+ opt->dict_size = value;
+ break;
+
+ case OPT_LC:
+ opt->lc = value;
+ break;
+
+ case OPT_LP:
+ opt->lp = value;
+ break;
+
+ case OPT_PB:
+ opt->pb = value;
+ break;
+
+ case OPT_MODE:
+ opt->mode = value;
+ break;
+
+ case OPT_NICE:
+ opt->nice_len = value;
+ break;
+
+ case OPT_MF:
+ opt->mf = value;
+ break;
+
+ case OPT_DEPTH:
+ opt->depth = value;
+ break;
+ }
+}
+
+
+extern lzma_options_lzma *
+options_lzma(const char *str)
+{
+ static const name_id_map modes[] = {
+ { "fast", LZMA_MODE_FAST },
+ { "normal", LZMA_MODE_NORMAL },
+ { NULL, 0 }
+ };
+
+ static const name_id_map mfs[] = {
+ { "hc3", LZMA_MF_HC3 },
+ { "hc4", LZMA_MF_HC4 },
+ { "bt2", LZMA_MF_BT2 },
+ { "bt3", LZMA_MF_BT3 },
+ { "bt4", LZMA_MF_BT4 },
+ { NULL, 0 }
+ };
+
+ static const option_map opts[] = {
+ { "preset", NULL, UINT64_MAX, 0 },
+ { "dict", NULL, LZMA_DICT_SIZE_MIN,
+ (UINT32_C(1) << 30) + (UINT32_C(1) << 29) },
+ { "lc", NULL, LZMA_LCLP_MIN, LZMA_LCLP_MAX },
+ { "lp", NULL, LZMA_LCLP_MIN, LZMA_LCLP_MAX },
+ { "pb", NULL, LZMA_PB_MIN, LZMA_PB_MAX },
+ { "mode", modes, 0, 0 },
+ { "nice", NULL, 2, 273 },
+ { "mf", mfs, 0, 0 },
+ { "depth", NULL, 0, UINT32_MAX },
+ { NULL, NULL, 0, 0 }
+ };
+
+ lzma_options_lzma *options = xmalloc(sizeof(lzma_options_lzma));
+ if (lzma_lzma_preset(options, LZMA_PRESET_DEFAULT))
+ message_bug();
+
+ parse_options(str, opts, &set_lzma, options);
+
+ if (options->lc + options->lp > LZMA_LCLP_MAX)
+ message_fatal(_("The sum of lc and lp must not exceed 4"));
+
+ return options;
+}
diff --git a/src/xz/options.h b/src/xz/options.h
new file mode 100644
index 0000000..61ec8d5
--- /dev/null
+++ b/src/xz/options.h
@@ -0,0 +1,31 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file options.h
+/// \brief Parser for filter-specific options
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// \brief Parser for Delta options
+///
+/// \return Pointer to allocated options structure.
+/// Doesn't return on error.
+extern lzma_options_delta *options_delta(const char *str);
+
+
+/// \brief Parser for BCJ options
+///
+/// \return Pointer to allocated options structure.
+/// Doesn't return on error.
+extern lzma_options_bcj *options_bcj(const char *str);
+
+
+/// \brief Parser for LZMA options
+///
+/// \return Pointer to allocated options structure.
+/// Doesn't return on error.
+extern lzma_options_lzma *options_lzma(const char *str);
diff --git a/src/xz/private.h b/src/xz/private.h
new file mode 100644
index 0000000..6414bdb
--- /dev/null
+++ b/src/xz/private.h
@@ -0,0 +1,66 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file private.h
+/// \brief Common includes, definitions, and prototypes
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include "mythread.h"
+
+#include "lzma.h"
+
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <errno.h>
+#include <signal.h>
+#include <locale.h>
+#include <stdio.h>
+#include <unistd.h>
+
+#include "tuklib_gettext.h"
+#include "tuklib_progname.h"
+#include "tuklib_exit.h"
+#include "tuklib_mbstr.h"
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+# define WIN32_LEAN_AND_MEAN
+# include <windows.h>
+#endif
+
+#ifndef STDIN_FILENO
+# define STDIN_FILENO (fileno(stdin))
+#endif
+
+#ifndef STDOUT_FILENO
+# define STDOUT_FILENO (fileno(stdout))
+#endif
+
+#ifndef STDERR_FILENO
+# define STDERR_FILENO (fileno(stderr))
+#endif
+
+#if defined(HAVE_CAPSICUM) || defined(HAVE_PLEDGE)
+# define ENABLE_SANDBOX 1
+#endif
+
+#include "main.h"
+#include "mytime.h"
+#include "coder.h"
+#include "message.h"
+#include "args.h"
+#include "hardware.h"
+#include "file_io.h"
+#include "options.h"
+#include "signals.h"
+#include "suffix.h"
+#include "util.h"
+
+#ifdef HAVE_DECODERS
+# include "list.h"
+#endif
diff --git a/src/xz/signals.c b/src/xz/signals.c
new file mode 100644
index 0000000..7aef463
--- /dev/null
+++ b/src/xz/signals.c
@@ -0,0 +1,209 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file signals.c
+/// \brief Handling signals to abort operation
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+
+volatile sig_atomic_t user_abort = false;
+
+
+#if !(defined(_WIN32) && !defined(__CYGWIN__))
+
+/// If we were interrupted by a signal, we store the signal number so that
+/// we can raise that signal to kill the program when all cleanups have
+/// been done.
+static volatile sig_atomic_t exit_signal = 0;
+
+/// Mask of signals for which we have established a signal handler to set
+/// user_abort to true.
+static sigset_t hooked_signals;
+
+/// True once signals_init() has finished. This is used to skip blocking
+/// signals (with uninitialized hooked_signals) if signals_block() and
+/// signals_unblock() are called before signals_init() has been called.
+static bool signals_are_initialized = false;
+
+/// signals_block() and signals_unblock() can be called recursively.
+static size_t signals_block_count = 0;
+
+
+static void
+signal_handler(int sig)
+{
+ exit_signal = sig;
+ user_abort = true;
+
+#ifndef TUKLIB_DOSLIKE
+ io_write_to_user_abort_pipe();
+#endif
+
+ return;
+}
+
+
+extern void
+signals_init(void)
+{
+ // List of signals for which we establish the signal handler.
+ static const int sigs[] = {
+ SIGINT,
+ SIGTERM,
+#ifdef SIGHUP
+ SIGHUP,
+#endif
+#ifdef SIGPIPE
+ SIGPIPE,
+#endif
+#ifdef SIGXCPU
+ SIGXCPU,
+#endif
+#ifdef SIGXFSZ
+ SIGXFSZ,
+#endif
+ };
+
+ // Mask of the signals for which we have established a signal handler.
+ sigemptyset(&hooked_signals);
+ for (size_t i = 0; i < ARRAY_SIZE(sigs); ++i)
+ sigaddset(&hooked_signals, sigs[i]);
+
+#ifdef SIGALRM
+ // Add also the signals from message.c to hooked_signals.
+ for (size_t i = 0; message_progress_sigs[i] != 0; ++i)
+ sigaddset(&hooked_signals, message_progress_sigs[i]);
+#endif
+
+ // Using "my_sa" because "sa" may conflict with a sockaddr variable
+ // from system headers on Solaris.
+ struct sigaction my_sa;
+
+ // All the signals that we handle we also blocked while the signal
+ // handler runs.
+ my_sa.sa_mask = hooked_signals;
+
+ // Don't set SA_RESTART, because we want EINTR so that we can check
+ // for user_abort and cleanup before exiting. We block the signals
+ // for which we have established a handler when we don't want EINTR.
+ my_sa.sa_flags = 0;
+ my_sa.sa_handler = &signal_handler;
+
+ for (size_t i = 0; i < ARRAY_SIZE(sigs); ++i) {
+ // If the parent process has left some signals ignored,
+ // we don't unignore them.
+ struct sigaction old;
+ if (sigaction(sigs[i], NULL, &old) == 0
+ && old.sa_handler == SIG_IGN)
+ continue;
+
+ // Establish the signal handler.
+ if (sigaction(sigs[i], &my_sa, NULL))
+ message_signal_handler();
+ }
+
+ signals_are_initialized = true;
+
+ return;
+}
+
+
+#ifndef __VMS
+extern void
+signals_block(void)
+{
+ if (signals_are_initialized) {
+ if (signals_block_count++ == 0) {
+ const int saved_errno = errno;
+ mythread_sigmask(SIG_BLOCK, &hooked_signals, NULL);
+ errno = saved_errno;
+ }
+ }
+
+ return;
+}
+
+
+extern void
+signals_unblock(void)
+{
+ if (signals_are_initialized) {
+ assert(signals_block_count > 0);
+
+ if (--signals_block_count == 0) {
+ const int saved_errno = errno;
+ mythread_sigmask(SIG_UNBLOCK, &hooked_signals, NULL);
+ errno = saved_errno;
+ }
+ }
+
+ return;
+}
+#endif
+
+
+extern void
+signals_exit(void)
+{
+ const int sig = (int)exit_signal;
+
+ if (sig != 0) {
+#if defined(TUKLIB_DOSLIKE) || defined(__VMS)
+ // Don't raise(), set only exit status. This avoids
+ // printing unwanted message about SIGINT when the user
+ // presses C-c.
+ set_exit_status(E_ERROR);
+#else
+ struct sigaction sa;
+ sa.sa_handler = SIG_DFL;
+ sigfillset(&sa.sa_mask);
+ sa.sa_flags = 0;
+ sigaction(sig, &sa, NULL);
+ raise(sig);
+#endif
+ }
+
+ return;
+}
+
+#else
+
+// While Windows has some very basic signal handling functions as required
+// by C89, they are not really used, and e.g. SIGINT doesn't work exactly
+// the way it does on POSIX (Windows creates a new thread for the signal
+// handler). Instead, we use SetConsoleCtrlHandler() to catch user
+// pressing C-c, because that seems to be the recommended way to do it.
+//
+// NOTE: This doesn't work under MSYS. Trying with SIGINT doesn't work
+// either even if it appeared to work at first. So test using Windows
+// console window.
+
+static BOOL WINAPI
+signal_handler(DWORD type lzma_attribute((__unused__)))
+{
+ // Since we don't get a signal number which we could raise() at
+ // signals_exit() like on POSIX, just set the exit status to
+ // indicate an error, so that we cannot return with zero exit status.
+ set_exit_status(E_ERROR);
+ user_abort = true;
+ return TRUE;
+}
+
+
+extern void
+signals_init(void)
+{
+ if (!SetConsoleCtrlHandler(&signal_handler, TRUE))
+ message_signal_handler();
+
+ return;
+}
+
+#endif
diff --git a/src/xz/signals.h b/src/xz/signals.h
new file mode 100644
index 0000000..5b125e0
--- /dev/null
+++ b/src/xz/signals.h
@@ -0,0 +1,43 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file signals.h
+/// \brief Handling signals to abort operation
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// If this is true, we will clean up the possibly incomplete output file,
+/// return to main() as soon as practical. That is, the code needs to poll
+/// this variable in various places.
+extern volatile sig_atomic_t user_abort;
+
+
+/// Initialize the signal handler, which will set user_abort to true when
+/// user e.g. presses C-c.
+extern void signals_init(void);
+
+
+#if (defined(_WIN32) && !defined(__CYGWIN__)) || defined(__VMS)
+# define signals_block() do { } while (0)
+# define signals_unblock() do { } while (0)
+#else
+/// Block the signals which don't have SA_RESTART and which would just set
+/// user_abort to true. This is handy when we don't want to handle EINTR
+/// and don't want SA_RESTART either.
+extern void signals_block(void);
+
+/// Unblock the signals blocked by signals_block().
+extern void signals_unblock(void);
+#endif
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+# define signals_exit() do { } while (0)
+#else
+/// If user has sent us a signal earlier to terminate the process,
+/// re-raise that signal to actually terminate the process.
+extern void signals_exit(void);
+#endif
diff --git a/src/xz/suffix.c b/src/xz/suffix.c
new file mode 100644
index 0000000..6ce9787
--- /dev/null
+++ b/src/xz/suffix.c
@@ -0,0 +1,401 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file suffix.c
+/// \brief Checks filename suffix and creates the destination filename
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+#ifdef __DJGPP__
+# include <fcntl.h>
+#endif
+
+// For case-insensitive filename suffix on case-insensitive systems
+#if defined(TUKLIB_DOSLIKE) || defined(__VMS)
+# ifdef HAVE_STRINGS_H
+# include <strings.h>
+# endif
+# define strcmp strcasecmp
+#endif
+
+
+static char *custom_suffix = NULL;
+
+
+/// \brief Test if the char is a directory separator
+static bool
+is_dir_sep(char c)
+{
+#ifdef TUKLIB_DOSLIKE
+ return c == '/' || c == '\\' || c == ':';
+#else
+ return c == '/';
+#endif
+}
+
+
+/// \brief Test if the string contains a directory separator
+static bool
+has_dir_sep(const char *str)
+{
+#ifdef TUKLIB_DOSLIKE
+ return strpbrk(str, "/\\:") != NULL;
+#else
+ return strchr(str, '/') != NULL;
+#endif
+}
+
+
+#ifdef __DJGPP__
+/// \brief Test for special suffix used for 8.3 short filenames (SFN)
+///
+/// \return If str matches *.?- or *.??-, true is returned. Otherwise
+/// false is returned.
+static bool
+has_sfn_suffix(const char *str, size_t len)
+{
+ if (len >= 4 && str[len - 1] == '-' && str[len - 2] != '.'
+ && !is_dir_sep(str[len - 2])) {
+ // *.?-
+ if (str[len - 3] == '.')
+ return !is_dir_sep(str[len - 4]);
+
+ // *.??-
+ if (len >= 5 && !is_dir_sep(str[len - 3])
+ && str[len - 4] == '.')
+ return !is_dir_sep(str[len - 5]);
+ }
+
+ return false;
+}
+#endif
+
+
+/// \brief Checks if src_name has given compressed_suffix
+///
+/// \param suffix Filename suffix to look for
+/// \param src_name Input filename
+/// \param src_len strlen(src_name)
+///
+/// \return If src_name has the suffix, src_len - strlen(suffix) is
+/// returned. It's always a positive integer. Otherwise zero
+/// is returned.
+static size_t
+test_suffix(const char *suffix, const char *src_name, size_t src_len)
+{
+ const size_t suffix_len = strlen(suffix);
+
+ // The filename must have at least one character in addition to
+ // the suffix. src_name may contain path to the filename, so we
+ // need to check for directory separator too.
+ if (src_len <= suffix_len
+ || is_dir_sep(src_name[src_len - suffix_len - 1]))
+ return 0;
+
+ if (strcmp(suffix, src_name + src_len - suffix_len) == 0)
+ return src_len - suffix_len;
+
+ return 0;
+}
+
+
+/// \brief Removes the filename suffix of the compressed file
+///
+/// \return Name of the uncompressed file, or NULL if file has unknown
+/// suffix.
+static char *
+uncompressed_name(const char *src_name, const size_t src_len)
+{
+ static const struct {
+ const char *compressed;
+ const char *uncompressed;
+ } suffixes[] = {
+ { ".xz", "" },
+ { ".txz", ".tar" }, // .txz abbreviation for .txt.gz is rare.
+ { ".lzma", "" },
+#ifdef __DJGPP__
+ { ".lzm", "" },
+#endif
+ { ".tlz", ".tar" }, // Both .tar.lzma and .tar.lz
+#ifdef HAVE_LZIP_DECODER
+ { ".lz", "" },
+#endif
+ };
+
+ const char *new_suffix = "";
+ size_t new_len = 0;
+
+ if (opt_format != FORMAT_RAW) {
+ for (size_t i = 0; i < ARRAY_SIZE(suffixes); ++i) {
+ new_len = test_suffix(suffixes[i].compressed,
+ src_name, src_len);
+ if (new_len != 0) {
+ new_suffix = suffixes[i].uncompressed;
+ break;
+ }
+ }
+
+#ifdef __DJGPP__
+ // Support also *.?- -> *.? and *.??- -> *.?? on DOS.
+ // This is done also when long filenames are available
+ // to keep it easy to decompress files created when
+ // long filename support wasn't available.
+ if (new_len == 0 && has_sfn_suffix(src_name, src_len)) {
+ new_suffix = "";
+ new_len = src_len - 1;
+ }
+#endif
+ }
+
+ if (new_len == 0 && custom_suffix != NULL)
+ new_len = test_suffix(custom_suffix, src_name, src_len);
+
+ if (new_len == 0) {
+ message_warning(_("%s: Filename has an unknown suffix, "
+ "skipping"), src_name);
+ return NULL;
+ }
+
+ const size_t new_suffix_len = strlen(new_suffix);
+ char *dest_name = xmalloc(new_len + new_suffix_len + 1);
+
+ memcpy(dest_name, src_name, new_len);
+ memcpy(dest_name + new_len, new_suffix, new_suffix_len);
+ dest_name[new_len + new_suffix_len] = '\0';
+
+ return dest_name;
+}
+
+
+/// This message is needed in multiple places in compressed_name(),
+/// so the message has been put into its own function.
+static void
+msg_suffix(const char *src_name, const char *suffix)
+{
+ message_warning(_("%s: File already has `%s' suffix, skipping"),
+ src_name, suffix);
+ return;
+}
+
+
+/// \brief Appends suffix to src_name
+///
+/// In contrast to uncompressed_name(), we check only suffixes that are valid
+/// for the specified file format.
+static char *
+compressed_name(const char *src_name, size_t src_len)
+{
+ // The order of these must match the order in args.h.
+ static const char *const all_suffixes[][4] = {
+ {
+ ".xz",
+ ".txz",
+ NULL
+ }, {
+ ".lzma",
+#ifdef __DJGPP__
+ ".lzm",
+#endif
+ ".tlz",
+ NULL
+#ifdef HAVE_LZIP_DECODER
+ // This is needed to keep the table indexing in sync with
+ // enum format_type from coder.h.
+ }, {
+/*
+ ".lz",
+*/
+ NULL
+#endif
+ }, {
+ // --format=raw requires specifying the suffix
+ // manually or using stdout.
+ NULL
+ }
+ };
+
+ // args.c ensures these.
+ assert(opt_format != FORMAT_AUTO);
+#ifdef HAVE_LZIP_DECODER
+ assert(opt_format != FORMAT_LZIP);
+#endif
+
+ const size_t format = opt_format - 1;
+ const char *const *suffixes = all_suffixes[format];
+
+ // Look for known filename suffixes and refuse to compress them.
+ for (size_t i = 0; suffixes[i] != NULL; ++i) {
+ if (test_suffix(suffixes[i], src_name, src_len) != 0) {
+ msg_suffix(src_name, suffixes[i]);
+ return NULL;
+ }
+ }
+
+#ifdef __DJGPP__
+ // Recognize also the special suffix that is used when long
+ // filename (LFN) support isn't available. This suffix is
+ // recognized on LFN systems too.
+ if (opt_format == FORMAT_XZ && has_sfn_suffix(src_name, src_len)) {
+ msg_suffix(src_name, "-");
+ return NULL;
+ }
+#endif
+
+ if (custom_suffix != NULL) {
+ if (test_suffix(custom_suffix, src_name, src_len) != 0) {
+ msg_suffix(src_name, custom_suffix);
+ return NULL;
+ }
+ }
+
+ const char *suffix = custom_suffix != NULL
+ ? custom_suffix : suffixes[0];
+ size_t suffix_len = strlen(suffix);
+
+#ifdef __DJGPP__
+ if (!_use_lfn(src_name)) {
+ // Long filename (LFN) support isn't available and we are
+ // limited to 8.3 short filenames (SFN).
+ //
+ // Look for suffix separator from the filename, and make sure
+ // that it is in the filename, not in a directory name.
+ const char *sufsep = strrchr(src_name, '.');
+ if (sufsep == NULL || sufsep[1] == '\0'
+ || has_dir_sep(sufsep)) {
+ // src_name has no filename extension.
+ //
+ // Examples:
+ // xz foo -> foo.xz
+ // xz -F lzma foo -> foo.lzm
+ // xz -S x foo -> foox
+ // xz -S x foo. -> foo.x
+ // xz -S x.y foo -> foox.y
+ // xz -S .x foo -> foo.x
+ // xz -S .x foo. -> foo.x
+ //
+ // Avoid double dots:
+ if (sufsep != NULL && sufsep[1] == '\0'
+ && suffix[0] == '.')
+ --src_len;
+
+ } else if (custom_suffix == NULL
+ && strcasecmp(sufsep, ".tar") == 0) {
+ // ".tar" is handled specially.
+ //
+ // Examples:
+ // xz foo.tar -> foo.txz
+ // xz -F lzma foo.tar -> foo.tlz
+ static const char *const tar_suffixes[] = {
+ ".txz", // .tar.xz
+ ".tlz", // .tar.lzma
+/*
+ ".tlz", // .tar.lz
+*/
+ };
+ suffix = tar_suffixes[format];
+ suffix_len = 4;
+ src_len -= 4;
+
+ } else {
+ if (custom_suffix == NULL && opt_format == FORMAT_XZ) {
+ // Instead of the .xz suffix, use a single
+ // character at the end of the filename
+ // extension. This is to minimize name
+ // conflicts when compressing multiple files
+ // with the same basename. E.g. foo.txt and
+ // foo.exe become foo.tx- and foo.ex-. Dash
+ // is rare as the last character of the
+ // filename extension, so it seems to be
+ // quite safe choice and it stands out better
+ // in directory listings than e.g. x. For
+ // comparison, gzip uses z.
+ suffix = "-";
+ suffix_len = 1;
+ }
+
+ if (suffix[0] == '.') {
+ // The first character of the suffix is a dot.
+ // Throw away the original filename extension
+ // and replace it with the new suffix.
+ //
+ // Examples:
+ // xz -F lzma foo.txt -> foo.lzm
+ // xz -S .x foo.txt -> foo.x
+ src_len = sufsep - src_name;
+
+ } else {
+ // The first character of the suffix is not
+ // a dot. Preserve the first 0-2 characters
+ // of the original filename extension.
+ //
+ // Examples:
+ // xz foo.txt -> foo.tx-
+ // xz -S x foo.c -> foo.cx
+ // xz -S ab foo.c -> foo.cab
+ // xz -S ab foo.txt -> foo.tab
+ // xz -S abc foo.txt -> foo.abc
+ //
+ // Truncate the suffix to three chars:
+ if (suffix_len > 3)
+ suffix_len = 3;
+
+ // If needed, overwrite 1-3 characters.
+ if (strlen(sufsep) > 4 - suffix_len)
+ src_len = sufsep - src_name
+ + 4 - suffix_len;
+ }
+ }
+ }
+#endif
+
+ char *dest_name = xmalloc(src_len + suffix_len + 1);
+
+ memcpy(dest_name, src_name, src_len);
+ memcpy(dest_name + src_len, suffix, suffix_len);
+ dest_name[src_len + suffix_len] = '\0';
+
+ return dest_name;
+}
+
+
+extern char *
+suffix_get_dest_name(const char *src_name)
+{
+ assert(src_name != NULL);
+
+ // Length of the name is needed in all cases to locate the end of
+ // the string to compare the suffix, so calculate the length here.
+ const size_t src_len = strlen(src_name);
+
+ return opt_mode == MODE_COMPRESS
+ ? compressed_name(src_name, src_len)
+ : uncompressed_name(src_name, src_len);
+}
+
+
+extern void
+suffix_set(const char *suffix)
+{
+ // Empty suffix and suffixes having a directory separator are
+ // rejected. Such suffixes would break things later.
+ if (suffix[0] == '\0' || has_dir_sep(suffix))
+ message_fatal(_("%s: Invalid filename suffix"), suffix);
+
+ // Replace the old custom_suffix (if any) with the new suffix.
+ free(custom_suffix);
+ custom_suffix = xstrdup(suffix);
+ return;
+}
+
+
+extern bool
+suffix_is_set(void)
+{
+ return custom_suffix != NULL;
+}
diff --git a/src/xz/suffix.h b/src/xz/suffix.h
new file mode 100644
index 0000000..135e905
--- /dev/null
+++ b/src/xz/suffix.h
@@ -0,0 +1,36 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file suffix.h
+/// \brief Checks filename suffix and creates the destination filename
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// \brief Get the name of the destination file
+///
+/// Depending on the global variable opt_mode, this tries to find a matching
+/// counterpart for src_name. If the name can be constructed, it is allocated
+/// and returned (caller must free it). On error, a message is printed and
+/// NULL is returned.
+extern char *suffix_get_dest_name(const char *src_name);
+
+
+/// \brief Set a custom filename suffix
+///
+/// This function calls xstrdup() for the given suffix, thus the caller
+/// doesn't need to keep the memory allocated. There can be only one custom
+/// suffix, thus if this is called multiple times, the old suffixes are freed
+/// and forgotten.
+extern void suffix_set(const char *suffix);
+
+
+/// \brief Check if a custom suffix has been set
+///
+/// Returns true if the internal tracking of the suffix string has been set
+/// and false if the string has not been set. This will keep the suffix
+/// string encapsulated instead of extern-ing the variable.
+extern bool suffix_is_set(void);
diff --git a/src/xz/util.c b/src/xz/util.c
new file mode 100644
index 0000000..6ab4c2d
--- /dev/null
+++ b/src/xz/util.c
@@ -0,0 +1,287 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file util.c
+/// \brief Miscellaneous utility functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+#include <stdarg.h>
+
+
+/// Buffers for uint64_to_str() and uint64_to_nicestr()
+static char bufs[4][128];
+
+
+// Thousand separator support in uint64_to_str() and uint64_to_nicestr():
+//
+// DJGPP 2.05 added support for thousands separators but it's broken
+// at least under WinXP with Finnish locale that uses a non-breaking space
+// as the thousands separator. Workaround by disabling thousands separators
+// for DJGPP builds.
+//
+// MSVC doesn't support thousand separators.
+#if defined(__DJGPP__) || defined(_MSC_VER)
+# define FORMAT_THOUSAND_SEP(prefix, suffix) prefix suffix
+# define check_thousand_sep(slot) do { } while (0)
+#else
+# define FORMAT_THOUSAND_SEP(prefix, suffix) ((thousand == WORKS) \
+ ? prefix "'" suffix \
+ : prefix suffix)
+
+static enum { UNKNOWN, WORKS, BROKEN } thousand = UNKNOWN;
+
+/// Check if thousands separator is supported. Run-time checking is easiest
+/// because it seems to be sometimes lacking even on a POSIXish system.
+/// Note that trying to use thousands separators when snprintf() doesn't
+/// support them results in undefined behavior. This just has happened to
+/// work well enough in practice.
+///
+/// This must be called before using the FORMAT_THOUSAND_SEP macro.
+static void
+check_thousand_sep(uint32_t slot)
+{
+ if (thousand == UNKNOWN) {
+ bufs[slot][0] = '\0';
+ snprintf(bufs[slot], sizeof(bufs[slot]), "%'u", 1U);
+ thousand = bufs[slot][0] == '1' ? WORKS : BROKEN;
+ }
+
+ return;
+}
+#endif
+
+
+extern void *
+xrealloc(void *ptr, size_t size)
+{
+ assert(size > 0);
+
+ // Save ptr so that we can free it if realloc fails.
+ // The point is that message_fatal ends up calling stdio functions
+ // which in some libc implementations might allocate memory from
+ // the heap. Freeing ptr improves the chances that there's free
+ // memory for stdio functions if they need it.
+ void *p = ptr;
+ ptr = realloc(ptr, size);
+
+ if (ptr == NULL) {
+ const int saved_errno = errno;
+ free(p);
+ message_fatal("%s", strerror(saved_errno));
+ }
+
+ return ptr;
+}
+
+
+extern char *
+xstrdup(const char *src)
+{
+ assert(src != NULL);
+ const size_t size = strlen(src) + 1;
+ char *dest = xmalloc(size);
+ return memcpy(dest, src, size);
+}
+
+
+extern uint64_t
+str_to_uint64(const char *name, const char *value, uint64_t min, uint64_t max)
+{
+ uint64_t result = 0;
+
+ // Skip blanks.
+ while (*value == ' ' || *value == '\t')
+ ++value;
+
+ // Accept special value "max". Supporting "min" doesn't seem useful.
+ if (strcmp(value, "max") == 0)
+ return max;
+
+ if (*value < '0' || *value > '9')
+ message_fatal(_("%s: Value is not a non-negative "
+ "decimal integer"), value);
+
+ do {
+ // Don't overflow.
+ if (result > UINT64_MAX / 10)
+ goto error;
+
+ result *= 10;
+
+ // Another overflow check
+ const uint32_t add = (uint32_t)(*value - '0');
+ if (UINT64_MAX - add < result)
+ goto error;
+
+ result += add;
+ ++value;
+ } while (*value >= '0' && *value <= '9');
+
+ if (*value != '\0') {
+ // Look for suffix. Originally this supported both base-2
+ // and base-10, but since there seems to be little need
+ // for base-10 in this program, treat everything as base-2
+ // and also be more relaxed about the case of the first
+ // letter of the suffix.
+ uint64_t multiplier = 0;
+ if (*value == 'k' || *value == 'K')
+ multiplier = UINT64_C(1) << 10;
+ else if (*value == 'm' || *value == 'M')
+ multiplier = UINT64_C(1) << 20;
+ else if (*value == 'g' || *value == 'G')
+ multiplier = UINT64_C(1) << 30;
+
+ ++value;
+
+ // Allow also e.g. Ki, KiB, and KB.
+ if (*value != '\0' && strcmp(value, "i") != 0
+ && strcmp(value, "iB") != 0
+ && strcmp(value, "B") != 0)
+ multiplier = 0;
+
+ if (multiplier == 0) {
+ message(V_ERROR, _("%s: Invalid multiplier suffix"),
+ value - 1);
+ message_fatal(_("Valid suffixes are `KiB' (2^10), "
+ "`MiB' (2^20), and `GiB' (2^30)."));
+ }
+
+ // Don't overflow here either.
+ if (result > UINT64_MAX / multiplier)
+ goto error;
+
+ result *= multiplier;
+ }
+
+ if (result < min || result > max)
+ goto error;
+
+ return result;
+
+error:
+ message_fatal(_("Value of the option `%s' must be in the range "
+ "[%" PRIu64 ", %" PRIu64 "]"),
+ name, min, max);
+}
+
+
+extern uint64_t
+round_up_to_mib(uint64_t n)
+{
+ return (n >> 20) + ((n & ((UINT32_C(1) << 20) - 1)) != 0);
+}
+
+
+extern const char *
+uint64_to_str(uint64_t value, uint32_t slot)
+{
+ assert(slot < ARRAY_SIZE(bufs));
+
+ check_thousand_sep(slot);
+
+ snprintf(bufs[slot], sizeof(bufs[slot]),
+ FORMAT_THOUSAND_SEP("%", PRIu64), value);
+
+ return bufs[slot];
+}
+
+
+extern const char *
+uint64_to_nicestr(uint64_t value, enum nicestr_unit unit_min,
+ enum nicestr_unit unit_max, bool always_also_bytes,
+ uint32_t slot)
+{
+ assert(unit_min <= unit_max);
+ assert(unit_max <= NICESTR_TIB);
+ assert(slot < ARRAY_SIZE(bufs));
+
+ check_thousand_sep(slot);
+
+ enum nicestr_unit unit = NICESTR_B;
+ char *pos = bufs[slot];
+ size_t left = sizeof(bufs[slot]);
+
+ if ((unit_min == NICESTR_B && value < 10000)
+ || unit_max == NICESTR_B) {
+ // The value is shown as bytes.
+ my_snprintf(&pos, &left, FORMAT_THOUSAND_SEP("%", "u"),
+ (unsigned int)value);
+ } else {
+ // Scale the value to a nicer unit. Unless unit_min and
+ // unit_max limit us, we will show at most five significant
+ // digits with one decimal place.
+ double d = (double)(value);
+ do {
+ d /= 1024.0;
+ ++unit;
+ } while (unit < unit_min || (d > 9999.9 && unit < unit_max));
+
+ my_snprintf(&pos, &left, FORMAT_THOUSAND_SEP("%", ".1f"), d);
+ }
+
+ static const char suffix[5][4] = { "B", "KiB", "MiB", "GiB", "TiB" };
+ my_snprintf(&pos, &left, " %s", suffix[unit]);
+
+ if (always_also_bytes && value >= 10000)
+ snprintf(pos, left, FORMAT_THOUSAND_SEP(" (%", PRIu64 " B)"),
+ value);
+
+ return bufs[slot];
+}
+
+
+extern void
+my_snprintf(char **pos, size_t *left, const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ const int len = vsnprintf(*pos, *left, fmt, ap);
+ va_end(ap);
+
+ // If an error occurred, we want the caller to think that the whole
+ // buffer was used. This way no more data will be written to the
+ // buffer. We don't need better error handling here, although it
+ // is possible that the result looks garbage on the terminal if
+ // e.g. an UTF-8 character gets split. That shouldn't (easily)
+ // happen though, because the buffers used have some extra room.
+ if (len < 0 || (size_t)(len) >= *left) {
+ *left = 0;
+ } else {
+ *pos += len;
+ *left -= (size_t)(len);
+ }
+
+ return;
+}
+
+
+extern bool
+is_tty_stdin(void)
+{
+ const bool ret = isatty(STDIN_FILENO);
+
+ if (ret)
+ message_error(_("Compressed data cannot be read from "
+ "a terminal"));
+
+ return ret;
+}
+
+
+extern bool
+is_tty_stdout(void)
+{
+ const bool ret = isatty(STDOUT_FILENO);
+
+ if (ret)
+ message_error(_("Compressed data cannot be written to "
+ "a terminal"));
+
+ return ret;
+}
diff --git a/src/xz/util.h b/src/xz/util.h
new file mode 100644
index 0000000..6d7e148
--- /dev/null
+++ b/src/xz/util.h
@@ -0,0 +1,119 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file util.h
+/// \brief Miscellaneous utility functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// \brief Safe malloc() that never returns NULL
+///
+/// \note xmalloc(), xrealloc(), and xstrdup() must not be used when
+/// there are files open for writing, that should be cleaned up
+/// before exiting.
+#define xmalloc(size) xrealloc(NULL, size)
+
+
+/// \brief Safe realloc() that never returns NULL
+lzma_attr_alloc_size(2)
+extern void *xrealloc(void *ptr, size_t size);
+
+
+/// \brief Safe strdup() that never returns NULL
+extern char *xstrdup(const char *src);
+
+
+/// \brief Fancy version of strtoull()
+///
+/// \param name Name of the option to show in case of an error
+/// \param value String containing the number to be parsed; may
+/// contain suffixes "k", "M", "G", "Ki", "Mi", or "Gi"
+/// \param min Minimum valid value
+/// \param max Maximum valid value
+///
+/// \return Parsed value that is in the range [min, max]. Does not return
+/// if an error occurs.
+///
+extern uint64_t str_to_uint64(const char *name, const char *value,
+ uint64_t min, uint64_t max);
+
+
+/// \brief Round an integer up to the next full MiB and convert to MiB
+///
+/// This is used when printing memory usage and limit.
+extern uint64_t round_up_to_mib(uint64_t n);
+
+
+/// \brief Convert uint64_t to a string
+///
+/// Convert the given value to a string with locale-specific thousand
+/// separators, if supported by the snprintf() implementation. The string
+/// is stored into an internal static buffer indicated by the slot argument.
+/// A pointer to the selected buffer is returned.
+///
+/// This function exists, because non-POSIX systems don't support thousand
+/// separator in format strings. Solving the problem in a simple way doesn't
+/// work, because it breaks gettext (specifically, the xgettext tool).
+extern const char *uint64_to_str(uint64_t value, uint32_t slot);
+
+
+enum nicestr_unit {
+ NICESTR_B,
+ NICESTR_KIB,
+ NICESTR_MIB,
+ NICESTR_GIB,
+ NICESTR_TIB,
+};
+
+
+/// \brief Convert uint64_t to a nice human readable string
+///
+/// This is like uint64_to_str() but uses B, KiB, MiB, GiB, or TiB suffix
+/// and optionally includes the exact size in parenthesis.
+///
+/// \param value Value to be printed
+/// \param unit_min Smallest unit to use. This and unit_max are used
+/// e.g. when showing the progress indicator to force
+/// the unit to MiB.
+/// \param unit_max Biggest unit to use. assert(unit_min <= unit_max).
+/// \param always_also_bytes
+/// Show also the exact byte value in parenthesis
+/// if the nicely formatted string uses bigger unit
+/// than bytes.
+/// \param slot Which static buffer to use to hold the string.
+/// This is shared with uint64_to_str().
+///
+/// \return Pointer to statically allocated buffer containing the string.
+///
+/// \note This uses double_to_str() internally so the static buffer
+/// in double_to_str() will be overwritten.
+///
+extern const char *uint64_to_nicestr(uint64_t value,
+ enum nicestr_unit unit_min, enum nicestr_unit unit_max,
+ bool always_also_bytes, uint32_t slot);
+
+
+/// \brief Wrapper for snprintf() to help constructing a string in pieces
+///
+/// A maximum of *left bytes is written starting from *pos. *pos and *left
+/// are updated accordingly.
+lzma_attribute((__format__(__printf__, 3, 4)))
+extern void my_snprintf(char **pos, size_t *left, const char *fmt, ...);
+
+
+/// \brief Test if stdin is a terminal
+///
+/// If stdin is a terminal, an error message is printed and exit status set
+/// to EXIT_ERROR.
+extern bool is_tty_stdin(void);
+
+
+/// \brief Test if stdout is a terminal
+///
+/// If stdout is a terminal, an error message is printed and exit status set
+/// to EXIT_ERROR.
+extern bool is_tty_stdout(void);
diff --git a/src/xz/xz.1 b/src/xz/xz.1
new file mode 100644
index 0000000..73ca6ef
--- /dev/null
+++ b/src/xz/xz.1
@@ -0,0 +1,3020 @@
+'\" t
+.\"
+.\" Authors: Lasse Collin
+.\" Jia Tan
+.\"
+.\" This file has been put into the public domain.
+.\" You can do whatever you want with this file.
+.\"
+.TH XZ 1 "2023-07-17" "Tukaani" "XZ Utils"
+.
+.SH NAME
+xz, unxz, xzcat, lzma, unlzma, lzcat \- Compress or decompress .xz and .lzma files
+.
+.SH SYNOPSIS
+.B xz
+.RI [ option... ]
+.RI [ file... ]
+.
+.SH COMMAND ALIASES
+.B unxz
+is equivalent to
+.BR "xz \-\-decompress" .
+.br
+.B xzcat
+is equivalent to
+.BR "xz \-\-decompress \-\-stdout" .
+.br
+.B lzma
+is equivalent to
+.BR "xz \-\-format=lzma" .
+.br
+.B unlzma
+is equivalent to
+.BR "xz \-\-format=lzma \-\-decompress" .
+.br
+.B lzcat
+is equivalent to
+.BR "xz \-\-format=lzma \-\-decompress \-\-stdout" .
+.PP
+When writing scripts that need to decompress files,
+it is recommended to always use the name
+.B xz
+with appropriate arguments
+.RB ( "xz \-d"
+or
+.BR "xz \-dc" )
+instead of the names
+.B unxz
+and
+.BR xzcat .
+.
+.SH DESCRIPTION
+.B xz
+is a general-purpose data compression tool with
+command line syntax similar to
+.BR gzip (1)
+and
+.BR bzip2 (1).
+The native file format is the
+.B .xz
+format, but the legacy
+.B .lzma
+format used by LZMA Utils and
+raw compressed streams with no container format headers
+are also supported.
+In addition, decompression of the
+.B .lz
+format used by
+.B lzip
+is supported.
+.PP
+.B xz
+compresses or decompresses each
+.I file
+according to the selected operation mode.
+If no
+.I files
+are given or
+.I file
+is
+.BR \- ,
+.B xz
+reads from standard input and writes the processed data
+to standard output.
+.B xz
+will refuse (display an error and skip the
+.IR file )
+to write compressed data to standard output if it is a terminal.
+Similarly,
+.B xz
+will refuse to read compressed data
+from standard input if it is a terminal.
+.PP
+Unless
+.B \-\-stdout
+is specified,
+.I files
+other than
+.B \-
+are written to a new file whose name is derived from the source
+.I file
+name:
+.IP \(bu 3
+When compressing, the suffix of the target file format
+.RB ( .xz
+or
+.BR .lzma )
+is appended to the source filename to get the target filename.
+.IP \(bu 3
+When decompressing, the
+.BR .xz ,
+.BR .lzma ,
+or
+.B .lz
+suffix is removed from the filename to get the target filename.
+.B xz
+also recognizes the suffixes
+.B .txz
+and
+.BR .tlz ,
+and replaces them with the
+.B .tar
+suffix.
+.PP
+If the target file already exists, an error is displayed and the
+.I file
+is skipped.
+.PP
+Unless writing to standard output,
+.B xz
+will display a warning and skip the
+.I file
+if any of the following applies:
+.IP \(bu 3
+.I File
+is not a regular file.
+Symbolic links are not followed,
+and thus they are not considered to be regular files.
+.IP \(bu 3
+.I File
+has more than one hard link.
+.IP \(bu 3
+.I File
+has setuid, setgid, or sticky bit set.
+.IP \(bu 3
+The operation mode is set to compress and the
+.I file
+already has a suffix of the target file format
+.RB ( .xz
+or
+.B .txz
+when compressing to the
+.B .xz
+format, and
+.B .lzma
+or
+.B .tlz
+when compressing to the
+.B .lzma
+format).
+.IP \(bu 3
+The operation mode is set to decompress and the
+.I file
+doesn't have a suffix of any of the supported file formats
+.RB ( .xz ,
+.BR .txz ,
+.BR .lzma ,
+.BR .tlz ,
+or
+.BR .lz ).
+.PP
+After successfully compressing or decompressing the
+.IR file ,
+.B xz
+copies the owner, group, permissions, access time,
+and modification time from the source
+.I file
+to the target file.
+If copying the group fails, the permissions are modified
+so that the target file doesn't become accessible to users
+who didn't have permission to access the source
+.IR file .
+.B xz
+doesn't support copying other metadata like access control lists
+or extended attributes yet.
+.PP
+Once the target file has been successfully closed, the source
+.I file
+is removed unless
+.B \-\-keep
+was specified.
+The source
+.I file
+is never removed if the output is written to standard output
+or if an error occurs.
+.PP
+Sending
+.B SIGINFO
+or
+.B SIGUSR1
+to the
+.B xz
+process makes it print progress information to standard error.
+This has only limited use since when standard error
+is a terminal, using
+.B \-\-verbose
+will display an automatically updating progress indicator.
+.
+.SS "Memory usage"
+The memory usage of
+.B xz
+varies from a few hundred kilobytes to several gigabytes
+depending on the compression settings.
+The settings used when compressing a file determine
+the memory requirements of the decompressor.
+Typically the decompressor needs 5\ % to 20\ % of
+the amount of memory that the compressor needed when
+creating the file.
+For example, decompressing a file created with
+.B xz \-9
+currently requires 65\ MiB of memory.
+Still, it is possible to have
+.B .xz
+files that require several gigabytes of memory to decompress.
+.PP
+Especially users of older systems may find
+the possibility of very large memory usage annoying.
+To prevent uncomfortable surprises,
+.B xz
+has a built-in memory usage limiter, which is disabled by default.
+While some operating systems provide ways to limit
+the memory usage of processes, relying on it
+wasn't deemed to be flexible enough (for example, using
+.BR ulimit (1)
+to limit virtual memory tends to cripple
+.BR mmap (2)).
+.PP
+The memory usage limiter can be enabled with
+the command line option \fB\-\-memlimit=\fIlimit\fR.
+Often it is more convenient to enable the limiter
+by default by setting the environment variable
+.BR XZ_DEFAULTS ,
+for example,
+.BR XZ_DEFAULTS=\-\-memlimit=150MiB .
+It is possible to set the limits separately
+for compression and decompression by using
+.BI \-\-memlimit\-compress= limit
+and \fB\-\-memlimit\-decompress=\fIlimit\fR.
+Using these two options outside
+.B XZ_DEFAULTS
+is rarely useful because a single run of
+.B xz
+cannot do both compression and decompression and
+.BI \-\-memlimit= limit
+(or
+.B \-M
+.IR limit )
+is shorter to type on the command line.
+.PP
+If the specified memory usage limit is exceeded when decompressing,
+.B xz
+will display an error and decompressing the file will fail.
+If the limit is exceeded when compressing,
+.B xz
+will try to scale the settings down so that the limit
+is no longer exceeded (except when using
+.B \-\-format=raw
+or
+.BR \-\-no\-adjust ).
+This way the operation won't fail unless the limit is very small.
+The scaling of the settings is done in steps that don't
+match the compression level presets, for example, if the limit is
+only slightly less than the amount required for
+.BR "xz \-9" ,
+the settings will be scaled down only a little,
+not all the way down to
+.BR "xz \-8" .
+.
+.SS "Concatenation and padding with .xz files"
+It is possible to concatenate
+.B .xz
+files as is.
+.B xz
+will decompress such files as if they were a single
+.B .xz
+file.
+.PP
+It is possible to insert padding between the concatenated parts
+or after the last part.
+The padding must consist of null bytes and the size
+of the padding must be a multiple of four bytes.
+This can be useful, for example, if the
+.B .xz
+file is stored on a medium that measures file sizes
+in 512-byte blocks.
+.PP
+Concatenation and padding are not allowed with
+.B .lzma
+files or raw streams.
+.
+.SH OPTIONS
+.
+.SS "Integer suffixes and special values"
+In most places where an integer argument is expected,
+an optional suffix is supported to easily indicate large integers.
+There must be no space between the integer and the suffix.
+.TP
+.B KiB
+Multiply the integer by 1,024 (2^10).
+.BR Ki ,
+.BR k ,
+.BR kB ,
+.BR K ,
+and
+.B KB
+are accepted as synonyms for
+.BR KiB .
+.TP
+.B MiB
+Multiply the integer by 1,048,576 (2^20).
+.BR Mi ,
+.BR m ,
+.BR M ,
+and
+.B MB
+are accepted as synonyms for
+.BR MiB .
+.TP
+.B GiB
+Multiply the integer by 1,073,741,824 (2^30).
+.BR Gi ,
+.BR g ,
+.BR G ,
+and
+.B GB
+are accepted as synonyms for
+.BR GiB .
+.PP
+The special value
+.B max
+can be used to indicate the maximum integer value
+supported by the option.
+.
+.SS "Operation mode"
+If multiple operation mode options are given,
+the last one takes effect.
+.TP
+.BR \-z ", " \-\-compress
+Compress.
+This is the default operation mode when no operation mode option
+is specified and no other operation mode is implied from
+the command name (for example,
+.B unxz
+implies
+.BR \-\-decompress ).
+.TP
+.BR \-d ", " \-\-decompress ", " \-\-uncompress
+Decompress.
+.TP
+.BR \-t ", " \-\-test
+Test the integrity of compressed
+.IR files .
+This option is equivalent to
+.B "\-\-decompress \-\-stdout"
+except that the decompressed data is discarded instead of being
+written to standard output.
+No files are created or removed.
+.TP
+.BR \-l ", " \-\-list
+Print information about compressed
+.IR files .
+No uncompressed output is produced,
+and no files are created or removed.
+In list mode, the program cannot read
+the compressed data from standard
+input or from other unseekable sources.
+.IP ""
+The default listing shows basic information about
+.IR files ,
+one file per line.
+To get more detailed information, use also the
+.B \-\-verbose
+option.
+For even more information, use
+.B \-\-verbose
+twice, but note that this may be slow, because getting all the extra
+information requires many seeks.
+The width of verbose output exceeds
+80 characters, so piping the output to, for example,
+.B "less\ \-S"
+may be convenient if the terminal isn't wide enough.
+.IP ""
+The exact output may vary between
+.B xz
+versions and different locales.
+For machine-readable output,
+.B \-\-robot \-\-list
+should be used.
+.
+.SS "Operation modifiers"
+.TP
+.BR \-k ", " \-\-keep
+Don't delete the input files.
+.IP ""
+Since
+.B xz
+5.2.6,
+this option also makes
+.B xz
+compress or decompress even if the input is
+a symbolic link to a regular file,
+has more than one hard link,
+or has the setuid, setgid, or sticky bit set.
+The setuid, setgid, and sticky bits are not copied
+to the target file.
+In earlier versions this was only done with
+.BR \-\-force .
+.TP
+.BR \-f ", " \-\-force
+This option has several effects:
+.RS
+.IP \(bu 3
+If the target file already exists,
+delete it before compressing or decompressing.
+.IP \(bu 3
+Compress or decompress even if the input is
+a symbolic link to a regular file,
+has more than one hard link,
+or has the setuid, setgid, or sticky bit set.
+The setuid, setgid, and sticky bits are not copied
+to the target file.
+.IP \(bu 3
+When used with
+.B \-\-decompress
+.B \-\-stdout
+and
+.B xz
+cannot recognize the type of the source file,
+copy the source file as is to standard output.
+This allows
+.B xzcat
+.B \-\-force
+to be used like
+.BR cat (1)
+for files that have not been compressed with
+.BR xz .
+Note that in future,
+.B xz
+might support new compressed file formats, which may make
+.B xz
+decompress more types of files instead of copying them as is to
+standard output.
+.BI \-\-format= format
+can be used to restrict
+.B xz
+to decompress only a single file format.
+.RE
+.TP
+.BR \-c ", " \-\-stdout ", " \-\-to\-stdout
+Write the compressed or decompressed data to
+standard output instead of a file.
+This implies
+.BR \-\-keep .
+.TP
+.B \-\-single\-stream
+Decompress only the first
+.B .xz
+stream, and
+silently ignore possible remaining input data following the stream.
+Normally such trailing garbage makes
+.B xz
+display an error.
+.IP ""
+.B xz
+never decompresses more than one stream from
+.B .lzma
+files or raw streams, but this option still makes
+.B xz
+ignore the possible trailing data after the
+.B .lzma
+file or raw stream.
+.IP ""
+This option has no effect if the operation mode is not
+.B \-\-decompress
+or
+.BR \-\-test .
+.TP
+.B \-\-no\-sparse
+Disable creation of sparse files.
+By default, if decompressing into a regular file,
+.B xz
+tries to make the file sparse if the decompressed data contains
+long sequences of binary zeros.
+It also works when writing to standard output
+as long as standard output is connected to a regular file
+and certain additional conditions are met to make it safe.
+Creating sparse files may save disk space and speed up
+the decompression by reducing the amount of disk I/O.
+.TP
+\fB\-S\fR \fI.suf\fR, \fB\-\-suffix=\fI.suf
+When compressing, use
+.I .suf
+as the suffix for the target file instead of
+.B .xz
+or
+.BR .lzma .
+If not writing to standard output and
+the source file already has the suffix
+.IR .suf ,
+a warning is displayed and the file is skipped.
+.IP ""
+When decompressing, recognize files with the suffix
+.I .suf
+in addition to files with the
+.BR .xz ,
+.BR .txz ,
+.BR .lzma ,
+.BR .tlz ,
+or
+.B .lz
+suffix.
+If the source file has the suffix
+.IR .suf ,
+the suffix is removed to get the target filename.
+.IP ""
+When compressing or decompressing raw streams
+.RB ( \-\-format=raw ),
+the suffix must always be specified unless
+writing to standard output,
+because there is no default suffix for raw streams.
+.TP
+\fB\-\-files\fR[\fB=\fIfile\fR]
+Read the filenames to process from
+.IR file ;
+if
+.I file
+is omitted, filenames are read from standard input.
+Filenames must be terminated with the newline character.
+A dash
+.RB ( \- )
+is taken as a regular filename; it doesn't mean standard input.
+If filenames are given also as command line arguments, they are
+processed before the filenames read from
+.IR file .
+.TP
+\fB\-\-files0\fR[\fB=\fIfile\fR]
+This is identical to \fB\-\-files\fR[\fB=\fIfile\fR] except
+that each filename must be terminated with the null character.
+.
+.SS "Basic file format and compression options"
+.TP
+\fB\-F\fR \fIformat\fR, \fB\-\-format=\fIformat
+Specify the file
+.I format
+to compress or decompress:
+.RS
+.TP
+.B auto
+This is the default.
+When compressing,
+.B auto
+is equivalent to
+.BR xz .
+When decompressing,
+the format of the input file is automatically detected.
+Note that raw streams (created with
+.BR \-\-format=raw )
+cannot be auto-detected.
+.TP
+.B xz
+Compress to the
+.B .xz
+file format, or accept only
+.B .xz
+files when decompressing.
+.TP
+.BR lzma ", " alone
+Compress to the legacy
+.B .lzma
+file format, or accept only
+.B .lzma
+files when decompressing.
+The alternative name
+.B alone
+is provided for backwards compatibility with LZMA Utils.
+.TP
+.B lzip
+Accept only
+.B .lz
+files when decompressing.
+Compression is not supported.
+.IP ""
+The
+.B .lz
+format version 0 and the unextended version 1 are supported.
+Version 0 files were produced by
+.B lzip
+1.3 and older.
+Such files aren't common but may be found from file archives
+as a few source packages were released in this format.
+People might have old personal files in this format too.
+Decompression support for the format version 0 was removed in
+.B lzip
+1.18.
+.IP ""
+.B lzip
+1.4 and later create files in the format version 1.
+The sync flush marker extension to the format version 1 was added in
+.B lzip
+1.6.
+This extension is rarely used and isn't supported by
+.B xz
+(diagnosed as corrupt input).
+.TP
+.B raw
+Compress or uncompress a raw stream (no headers).
+This is meant for advanced users only.
+To decode raw streams, you need use
+.B \-\-format=raw
+and explicitly specify the filter chain,
+which normally would have been stored in the container headers.
+.RE
+.TP
+\fB\-C\fR \fIcheck\fR, \fB\-\-check=\fIcheck
+Specify the type of the integrity check.
+The check is calculated from the uncompressed data and
+stored in the
+.B .xz
+file.
+This option has an effect only when compressing into the
+.B .xz
+format; the
+.B .lzma
+format doesn't support integrity checks.
+The integrity check (if any) is verified when the
+.B .xz
+file is decompressed.
+.IP ""
+Supported
+.I check
+types:
+.RS
+.TP
+.B none
+Don't calculate an integrity check at all.
+This is usually a bad idea.
+This can be useful when integrity of the data is verified
+by other means anyway.
+.TP
+.B crc32
+Calculate CRC32 using the polynomial from IEEE-802.3 (Ethernet).
+.TP
+.B crc64
+Calculate CRC64 using the polynomial from ECMA-182.
+This is the default, since it is slightly better than CRC32
+at detecting damaged files and the speed difference is negligible.
+.TP
+.B sha256
+Calculate SHA-256.
+This is somewhat slower than CRC32 and CRC64.
+.RE
+.IP ""
+Integrity of the
+.B .xz
+headers is always verified with CRC32.
+It is not possible to change or disable it.
+.TP
+.B \-\-ignore\-check
+Don't verify the integrity check of the compressed data when decompressing.
+The CRC32 values in the
+.B .xz
+headers will still be verified normally.
+.IP ""
+.B "Do not use this option unless you know what you are doing."
+Possible reasons to use this option:
+.RS
+.IP \(bu 3
+Trying to recover data from a corrupt .xz file.
+.IP \(bu 3
+Speeding up decompression.
+This matters mostly with SHA-256 or
+with files that have compressed extremely well.
+It's recommended to not use this option for this purpose
+unless the file integrity is verified externally in some other way.
+.RE
+.TP
+.BR \-0 " ... " \-9
+Select a compression preset level.
+The default is
+.BR \-6 .
+If multiple preset levels are specified,
+the last one takes effect.
+If a custom filter chain was already specified, setting
+a compression preset level clears the custom filter chain.
+.IP ""
+The differences between the presets are more significant than with
+.BR gzip (1)
+and
+.BR bzip2 (1).
+The selected compression settings determine
+the memory requirements of the decompressor,
+thus using a too high preset level might make it painful
+to decompress the file on an old system with little RAM.
+Specifically,
+.B "it's not a good idea to blindly use \-9 for everything"
+like it often is with
+.BR gzip (1)
+and
+.BR bzip2 (1).
+.RS
+.TP
+.BR "\-0" " ... " "\-3"
+These are somewhat fast presets.
+.B \-0
+is sometimes faster than
+.B "gzip \-9"
+while compressing much better.
+The higher ones often have speed comparable to
+.BR bzip2 (1)
+with comparable or better compression ratio,
+although the results
+depend a lot on the type of data being compressed.
+.TP
+.BR "\-4" " ... " "\-6"
+Good to very good compression while keeping
+decompressor memory usage reasonable even for old systems.
+.B \-6
+is the default, which is usually a good choice
+for distributing files that need to be decompressible
+even on systems with only 16\ MiB RAM.
+.RB ( \-5e
+or
+.B \-6e
+may be worth considering too.
+See
+.BR \-\-extreme .)
+.TP
+.B "\-7 ... \-9"
+These are like
+.B \-6
+but with higher compressor and decompressor memory requirements.
+These are useful only when compressing files bigger than
+8\ MiB, 16\ MiB, and 32\ MiB, respectively.
+.RE
+.IP ""
+On the same hardware, the decompression speed is approximately
+a constant number of bytes of compressed data per second.
+In other words, the better the compression,
+the faster the decompression will usually be.
+This also means that the amount of uncompressed output
+produced per second can vary a lot.
+.IP ""
+The following table summarises the features of the presets:
+.RS
+.RS
+.PP
+.TS
+tab(;);
+c c c c c
+n n n n n.
+Preset;DictSize;CompCPU;CompMem;DecMem
+\-0;256 KiB;0;3 MiB;1 MiB
+\-1;1 MiB;1;9 MiB;2 MiB
+\-2;2 MiB;2;17 MiB;3 MiB
+\-3;4 MiB;3;32 MiB;5 MiB
+\-4;4 MiB;4;48 MiB;5 MiB
+\-5;8 MiB;5;94 MiB;9 MiB
+\-6;8 MiB;6;94 MiB;9 MiB
+\-7;16 MiB;6;186 MiB;17 MiB
+\-8;32 MiB;6;370 MiB;33 MiB
+\-9;64 MiB;6;674 MiB;65 MiB
+.TE
+.RE
+.RE
+.IP ""
+Column descriptions:
+.RS
+.IP \(bu 3
+DictSize is the LZMA2 dictionary size.
+It is waste of memory to use a dictionary bigger than
+the size of the uncompressed file.
+This is why it is good to avoid using the presets
+.BR \-7 " ... " \-9
+when there's no real need for them.
+At
+.B \-6
+and lower, the amount of memory wasted is
+usually low enough to not matter.
+.IP \(bu 3
+CompCPU is a simplified representation of the LZMA2 settings
+that affect compression speed.
+The dictionary size affects speed too,
+so while CompCPU is the same for levels
+.BR \-6 " ... " \-9 ,
+higher levels still tend to be a little slower.
+To get even slower and thus possibly better compression, see
+.BR \-\-extreme .
+.IP \(bu 3
+CompMem contains the compressor memory requirements
+in the single-threaded mode.
+It may vary slightly between
+.B xz
+versions.
+Memory requirements of some of the future multithreaded modes may
+be dramatically higher than that of the single-threaded mode.
+.IP \(bu 3
+DecMem contains the decompressor memory requirements.
+That is, the compression settings determine
+the memory requirements of the decompressor.
+The exact decompressor memory usage is slightly more than
+the LZMA2 dictionary size, but the values in the table
+have been rounded up to the next full MiB.
+.RE
+.TP
+.BR \-e ", " \-\-extreme
+Use a slower variant of the selected compression preset level
+.RB ( \-0 " ... " \-9 )
+to hopefully get a little bit better compression ratio,
+but with bad luck this can also make it worse.
+Decompressor memory usage is not affected,
+but compressor memory usage increases a little at preset levels
+.BR \-0 " ... " \-3 .
+.IP ""
+Since there are two presets with dictionary sizes
+4\ MiB and 8\ MiB, the presets
+.B \-3e
+and
+.B \-5e
+use slightly faster settings (lower CompCPU) than
+.B \-4e
+and
+.BR \-6e ,
+respectively.
+That way no two presets are identical.
+.RS
+.RS
+.PP
+.TS
+tab(;);
+c c c c c
+n n n n n.
+Preset;DictSize;CompCPU;CompMem;DecMem
+\-0e;256 KiB;8;4 MiB;1 MiB
+\-1e;1 MiB;8;13 MiB;2 MiB
+\-2e;2 MiB;8;25 MiB;3 MiB
+\-3e;4 MiB;7;48 MiB;5 MiB
+\-4e;4 MiB;8;48 MiB;5 MiB
+\-5e;8 MiB;7;94 MiB;9 MiB
+\-6e;8 MiB;8;94 MiB;9 MiB
+\-7e;16 MiB;8;186 MiB;17 MiB
+\-8e;32 MiB;8;370 MiB;33 MiB
+\-9e;64 MiB;8;674 MiB;65 MiB
+.TE
+.RE
+.RE
+.IP ""
+For example, there are a total of four presets that use
+8\ MiB dictionary, whose order from the fastest to the slowest is
+.BR \-5 ,
+.BR \-6 ,
+.BR \-5e ,
+and
+.BR \-6e .
+.TP
+.B \-\-fast
+.PD 0
+.TP
+.B \-\-best
+.PD
+These are somewhat misleading aliases for
+.B \-0
+and
+.BR \-9 ,
+respectively.
+These are provided only for backwards compatibility
+with LZMA Utils.
+Avoid using these options.
+.TP
+.BI \-\-block\-size= size
+When compressing to the
+.B .xz
+format, split the input data into blocks of
+.I size
+bytes.
+The blocks are compressed independently from each other,
+which helps with multi-threading and
+makes limited random-access decompression possible.
+This option is typically used to override the default
+block size in multi-threaded mode,
+but this option can be used in single-threaded mode too.
+.IP ""
+In multi-threaded mode about three times
+.I size
+bytes will be allocated in each thread for buffering input and output.
+The default
+.I size
+is three times the LZMA2 dictionary size or 1 MiB,
+whichever is more.
+Typically a good value is 2\(en4 times
+the size of the LZMA2 dictionary or at least 1 MiB.
+Using
+.I size
+less than the LZMA2 dictionary size is waste of RAM
+because then the LZMA2 dictionary buffer will never get fully used.
+The sizes of the blocks are stored in the block headers,
+which a future version of
+.B xz
+will use for multi-threaded decompression.
+.IP ""
+In single-threaded mode no block splitting is done by default.
+Setting this option doesn't affect memory usage.
+No size information is stored in block headers,
+thus files created in single-threaded mode
+won't be identical to files created in multi-threaded mode.
+The lack of size information also means that a future version of
+.B xz
+won't be able decompress the files in multi-threaded mode.
+.TP
+.BI \-\-block\-list= sizes
+When compressing to the
+.B .xz
+format, start a new block after
+the given intervals of uncompressed data.
+.IP ""
+The uncompressed
+.I sizes
+of the blocks are specified as a comma-separated list.
+Omitting a size (two or more consecutive commas) is a shorthand
+to use the size of the previous block.
+.IP ""
+If the input file is bigger than the sum of
+.IR sizes ,
+the last value in
+.I sizes
+is repeated until the end of the file.
+A special value of
+.B 0
+may be used as the last value to indicate that
+the rest of the file should be encoded as a single block.
+.IP ""
+If one specifies
+.I sizes
+that exceed the encoder's block size
+(either the default value in threaded mode or
+the value specified with \fB\-\-block\-size=\fIsize\fR),
+the encoder will create additional blocks while
+keeping the boundaries specified in
+.IR sizes .
+For example, if one specifies
+.B \-\-block\-size=10MiB
+.B \-\-block\-list=5MiB,10MiB,8MiB,12MiB,24MiB
+and the input file is 80 MiB,
+one will get 11 blocks:
+5, 10, 8, 10, 2, 10, 10, 4, 10, 10, and 1 MiB.
+.IP ""
+In multi-threaded mode the sizes of the blocks
+are stored in the block headers.
+This isn't done in single-threaded mode,
+so the encoded output won't be
+identical to that of the multi-threaded mode.
+.TP
+.BI \-\-flush\-timeout= timeout
+When compressing, if more than
+.I timeout
+milliseconds (a positive integer) has passed since the previous flush and
+reading more input would block,
+all the pending input data is flushed from the encoder and
+made available in the output stream.
+This can be useful if
+.B xz
+is used to compress data that is streamed over a network.
+Small
+.I timeout
+values make the data available at the receiving end
+with a small delay, but large
+.I timeout
+values give better compression ratio.
+.IP ""
+This feature is disabled by default.
+If this option is specified more than once, the last one takes effect.
+The special
+.I timeout
+value of
+.B 0
+can be used to explicitly disable this feature.
+.IP ""
+This feature is not available on non-POSIX systems.
+.IP ""
+.\" FIXME
+.B "This feature is still experimental."
+Currently
+.B xz
+is unsuitable for decompressing the stream in real time due to how
+.B xz
+does buffering.
+.TP
+.BI \-\-memlimit\-compress= limit
+Set a memory usage limit for compression.
+If this option is specified multiple times,
+the last one takes effect.
+.IP ""
+If the compression settings exceed the
+.IR limit ,
+.B xz
+will attempt to adjust the settings downwards so that
+the limit is no longer exceeded and display a notice that
+automatic adjustment was done.
+The adjustments are done in this order:
+reducing the number of threads,
+switching to single-threaded mode
+if even one thread in multi-threaded mode exceeds the
+.IR limit ,
+and finally reducing the LZMA2 dictionary size.
+.IP ""
+When compressing with
+.B \-\-format=raw
+or if
+.B \-\-no\-adjust
+has been specified,
+only the number of threads may be reduced
+since it can be done without affecting the compressed output.
+.IP ""
+If the
+.I limit
+cannot be met even with the adjustments described above,
+an error is displayed and
+.B xz
+will exit with exit status 1.
+.IP ""
+The
+.I limit
+can be specified in multiple ways:
+.RS
+.IP \(bu 3
+The
+.I limit
+can be an absolute value in bytes.
+Using an integer suffix like
+.B MiB
+can be useful.
+Example:
+.B "\-\-memlimit\-compress=80MiB"
+.IP \(bu 3
+The
+.I limit
+can be specified as a percentage of total physical memory (RAM).
+This can be useful especially when setting the
+.B XZ_DEFAULTS
+environment variable in a shell initialization script
+that is shared between different computers.
+That way the limit is automatically bigger
+on systems with more memory.
+Example:
+.B "\-\-memlimit\-compress=70%"
+.IP \(bu 3
+The
+.I limit
+can be reset back to its default value by setting it to
+.BR 0 .
+This is currently equivalent to setting the
+.I limit
+to
+.B max
+(no memory usage limit).
+.RE
+.IP ""
+For 32-bit
+.B xz
+there is a special case: if the
+.I limit
+would be over
+.BR "4020\ MiB" ,
+the
+.I limit
+is set to
+.BR "4020\ MiB" .
+On MIPS32
+.B "2000\ MiB"
+is used instead.
+(The values
+.B 0
+and
+.B max
+aren't affected by this.
+A similar feature doesn't exist for decompression.)
+This can be helpful when a 32-bit executable has access
+to 4\ GiB address space (2 GiB on MIPS32)
+while hopefully doing no harm in other situations.
+.IP ""
+See also the section
+.BR "Memory usage" .
+.TP
+.BI \-\-memlimit\-decompress= limit
+Set a memory usage limit for decompression.
+This also affects the
+.B \-\-list
+mode.
+If the operation is not possible without exceeding the
+.IR limit ,
+.B xz
+will display an error and decompressing the file will fail.
+See
+.BI \-\-memlimit\-compress= limit
+for possible ways to specify the
+.IR limit .
+.TP
+.BI \-\-memlimit\-mt\-decompress= limit
+Set a memory usage limit for multi-threaded decompression.
+This can only affect the number of threads;
+this will never make
+.B xz
+refuse to decompress a file.
+If
+.I limit
+is too low to allow any multi-threading, the
+.I limit
+is ignored and
+.B xz
+will continue in single-threaded mode.
+Note that if also
+.B \-\-memlimit\-decompress
+is used,
+it will always apply to both single-threaded and multi-threaded modes,
+and so the effective
+.I limit
+for multi-threading will never be higher than the limit set with
+.BR \-\-memlimit\-decompress .
+.IP ""
+In contrast to the other memory usage limit options,
+.BI \-\-memlimit\-mt\-decompress= limit
+has a system-specific default
+.IR limit .
+.B "xz \-\-info\-memory"
+can be used to see the current value.
+.IP ""
+This option and its default value exist
+because without any limit the threaded decompressor
+could end up allocating an insane amount of memory with some input files.
+If the default
+.I limit
+is too low on your system,
+feel free to increase the
+.I limit
+but never set it to a value larger than the amount of usable RAM
+as with appropriate input files
+.B xz
+will attempt to use that amount of memory
+even with a low number of threads.
+Running out of memory or swapping
+will not improve decompression performance.
+.IP ""
+See
+.BI \-\-memlimit\-compress= limit
+for possible ways to specify the
+.IR limit .
+Setting
+.I limit
+to
+.B 0
+resets the
+.I limit
+to the default system-specific value.
+.TP
+\fB\-M\fR \fIlimit\fR, \fB\-\-memlimit=\fIlimit\fR, \fB\-\-memory=\fIlimit
+This is equivalent to specifying
+.BI \-\-memlimit\-compress= limit
+.BI \-\-memlimit-decompress= limit
+\fB\-\-memlimit\-mt\-decompress=\fIlimit\fR.
+.TP
+.B \-\-no\-adjust
+Display an error and exit if the memory usage limit cannot be
+met without adjusting settings that affect the compressed output.
+That is, this prevents
+.B xz
+from switching the encoder from multi-threaded mode to single-threaded mode
+and from reducing the LZMA2 dictionary size.
+Even when this option is used the number of threads may be reduced
+to meet the memory usage limit as that won't affect the compressed output.
+.IP ""
+Automatic adjusting is always disabled when creating raw streams
+.RB ( \-\-format=raw ).
+.TP
+\fB\-T\fR \fIthreads\fR, \fB\-\-threads=\fIthreads
+Specify the number of worker threads to use.
+Setting
+.I threads
+to a special value
+.B 0
+makes
+.B xz
+use up to as many threads as the processor(s) on the system support.
+The actual number of threads can be fewer than
+.I threads
+if the input file is not big enough
+for threading with the given settings or
+if using more threads would exceed the memory usage limit.
+.IP ""
+The single-threaded and multi-threaded compressors produce different output.
+Single-threaded compressor will give the smallest file size but
+only the output from the multi-threaded compressor can be decompressed
+using multiple threads.
+Setting
+.I threads
+to
+.B 1
+will use the single-threaded mode.
+Setting
+.I threads
+to any other value, including
+.BR 0 ,
+will use the multi-threaded compressor
+even if the system supports only one hardware thread.
+.RB ( xz
+5.2.x
+used single-threaded mode in this situation.)
+.IP ""
+To use multi-threaded mode with only one thread, set
+.I threads
+to
+.BR +1 .
+The
+.B +
+prefix has no effect with values other than
+.BR 1 .
+A memory usage limit can still make
+.B xz
+switch to single-threaded mode unless
+.B \-\-no\-adjust
+is used.
+Support for the
+.B +
+prefix was added in
+.B xz
+5.4.0.
+.IP ""
+If an automatic number of threads has been requested and
+no memory usage limit has been specified,
+then a system-specific default soft limit will be used to possibly
+limit the number of threads.
+It is a soft limit in sense that it is ignored
+if the number of threads becomes one,
+thus a soft limit will never stop
+.B xz
+from compressing or decompressing.
+This default soft limit will not make
+.B xz
+switch from multi-threaded mode to single-threaded mode.
+The active limits can be seen with
+.BR "xz \-\-info\-memory" .
+.IP ""
+Currently the only threading method is to split the input into
+blocks and compress them independently from each other.
+The default block size depends on the compression level and
+can be overridden with the
+.BI \-\-block\-size= size
+option.
+.IP ""
+Threaded decompression only works on files that contain
+multiple blocks with size information in block headers.
+All large enough files compressed in multi-threaded mode
+meet this condition,
+but files compressed in single-threaded mode don't even if
+.BI \-\-block\-size= size
+has been used.
+.
+.SS "Custom compressor filter chains"
+A custom filter chain allows specifying
+the compression settings in detail instead of relying on
+the settings associated to the presets.
+When a custom filter chain is specified,
+preset options
+.RB ( \-0
+\&...\&
+.B \-9
+and
+.BR \-\-extreme )
+earlier on the command line are forgotten.
+If a preset option is specified
+after one or more custom filter chain options,
+the new preset takes effect and
+the custom filter chain options specified earlier are forgotten.
+.PP
+A filter chain is comparable to piping on the command line.
+When compressing, the uncompressed input goes to the first filter,
+whose output goes to the next filter (if any).
+The output of the last filter gets written to the compressed file.
+The maximum number of filters in the chain is four,
+but typically a filter chain has only one or two filters.
+.PP
+Many filters have limitations on where they can be
+in the filter chain:
+some filters can work only as the last filter in the chain,
+some only as a non-last filter, and some work in any position
+in the chain.
+Depending on the filter, this limitation is either inherent to
+the filter design or exists to prevent security issues.
+.PP
+A custom filter chain is specified by using one or more
+filter options in the order they are wanted in the filter chain.
+That is, the order of filter options is significant!
+When decoding raw streams
+.RB ( \-\-format=raw ),
+the filter chain is specified in the same order as
+it was specified when compressing.
+.PP
+Filters take filter-specific
+.I options
+as a comma-separated list.
+Extra commas in
+.I options
+are ignored.
+Every option has a default value, so you need to
+specify only those you want to change.
+.PP
+To see the whole filter chain and
+.IR options ,
+use
+.B "xz \-vv"
+(that is, use
+.B \-\-verbose
+twice).
+This works also for viewing the filter chain options used by presets.
+.TP
+\fB\-\-lzma1\fR[\fB=\fIoptions\fR]
+.PD 0
+.TP
+\fB\-\-lzma2\fR[\fB=\fIoptions\fR]
+.PD
+Add LZMA1 or LZMA2 filter to the filter chain.
+These filters can be used only as the last filter in the chain.
+.IP ""
+LZMA1 is a legacy filter,
+which is supported almost solely due to the legacy
+.B .lzma
+file format, which supports only LZMA1.
+LZMA2 is an updated
+version of LZMA1 to fix some practical issues of LZMA1.
+The
+.B .xz
+format uses LZMA2 and doesn't support LZMA1 at all.
+Compression speed and ratios of LZMA1 and LZMA2
+are practically the same.
+.IP ""
+LZMA1 and LZMA2 share the same set of
+.IR options :
+.RS
+.TP
+.BI preset= preset
+Reset all LZMA1 or LZMA2
+.I options
+to
+.IR preset .
+.I Preset
+consist of an integer, which may be followed by single-letter
+preset modifiers.
+The integer can be from
+.B 0
+to
+.BR 9 ,
+matching the command line options
+.B \-0
+\&...\&
+.BR \-9 .
+The only supported modifier is currently
+.BR e ,
+which matches
+.BR \-\-extreme .
+If no
+.B preset
+is specified, the default values of LZMA1 or LZMA2
+.I options
+are taken from the preset
+.BR 6 .
+.TP
+.BI dict= size
+Dictionary (history buffer)
+.I size
+indicates how many bytes of the recently processed
+uncompressed data is kept in memory.
+The algorithm tries to find repeating byte sequences (matches) in
+the uncompressed data, and replace them with references
+to the data currently in the dictionary.
+The bigger the dictionary, the higher is the chance
+to find a match.
+Thus, increasing dictionary
+.I size
+usually improves compression ratio, but
+a dictionary bigger than the uncompressed file is waste of memory.
+.IP ""
+Typical dictionary
+.I size
+is from 64\ KiB to 64\ MiB.
+The minimum is 4\ KiB.
+The maximum for compression is currently 1.5\ GiB (1536\ MiB).
+The decompressor already supports dictionaries up to
+one byte less than 4\ GiB, which is the maximum for
+the LZMA1 and LZMA2 stream formats.
+.IP ""
+Dictionary
+.I size
+and match finder
+.RI ( mf )
+together determine the memory usage of the LZMA1 or LZMA2 encoder.
+The same (or bigger) dictionary
+.I size
+is required for decompressing that was used when compressing,
+thus the memory usage of the decoder is determined
+by the dictionary size used when compressing.
+The
+.B .xz
+headers store the dictionary
+.I size
+either as
+.RI "2^" n
+or
+.RI "2^" n " + 2^(" n "\-1),"
+so these
+.I sizes
+are somewhat preferred for compression.
+Other
+.I sizes
+will get rounded up when stored in the
+.B .xz
+headers.
+.TP
+.BI lc= lc
+Specify the number of literal context bits.
+The minimum is 0 and the maximum is 4; the default is 3.
+In addition, the sum of
+.I lc
+and
+.I lp
+must not exceed 4.
+.IP ""
+All bytes that cannot be encoded as matches
+are encoded as literals.
+That is, literals are simply 8-bit bytes
+that are encoded one at a time.
+.IP ""
+The literal coding makes an assumption that the highest
+.I lc
+bits of the previous uncompressed byte correlate
+with the next byte.
+For example, in typical English text, an upper-case letter is
+often followed by a lower-case letter, and a lower-case
+letter is usually followed by another lower-case letter.
+In the US-ASCII character set, the highest three bits are 010
+for upper-case letters and 011 for lower-case letters.
+When
+.I lc
+is at least 3, the literal coding can take advantage of
+this property in the uncompressed data.
+.IP ""
+The default value (3) is usually good.
+If you want maximum compression, test
+.BR lc=4 .
+Sometimes it helps a little, and
+sometimes it makes compression worse.
+If it makes it worse, test
+.B lc=2
+too.
+.TP
+.BI lp= lp
+Specify the number of literal position bits.
+The minimum is 0 and the maximum is 4; the default is 0.
+.IP ""
+.I Lp
+affects what kind of alignment in the uncompressed data is
+assumed when encoding literals.
+See
+.I pb
+below for more information about alignment.
+.TP
+.BI pb= pb
+Specify the number of position bits.
+The minimum is 0 and the maximum is 4; the default is 2.
+.IP ""
+.I Pb
+affects what kind of alignment in the uncompressed data is
+assumed in general.
+The default means four-byte alignment
+.RI (2^ pb =2^2=4),
+which is often a good choice when there's no better guess.
+.IP ""
+When the alignment is known, setting
+.I pb
+accordingly may reduce the file size a little.
+For example, with text files having one-byte
+alignment (US-ASCII, ISO-8859-*, UTF-8), setting
+.B pb=0
+can improve compression slightly.
+For UTF-16 text,
+.B pb=1
+is a good choice.
+If the alignment is an odd number like 3 bytes,
+.B pb=0
+might be the best choice.
+.IP ""
+Even though the assumed alignment can be adjusted with
+.I pb
+and
+.IR lp ,
+LZMA1 and LZMA2 still slightly favor 16-byte alignment.
+It might be worth taking into account when designing file formats
+that are likely to be often compressed with LZMA1 or LZMA2.
+.TP
+.BI mf= mf
+Match finder has a major effect on encoder speed,
+memory usage, and compression ratio.
+Usually Hash Chain match finders are faster than Binary Tree
+match finders.
+The default depends on the
+.IR preset :
+0 uses
+.BR hc3 ,
+1\(en3
+use
+.BR hc4 ,
+and the rest use
+.BR bt4 .
+.IP ""
+The following match finders are supported.
+The memory usage formulas below are rough approximations,
+which are closest to the reality when
+.I dict
+is a power of two.
+.RS
+.TP
+.B hc3
+Hash Chain with 2- and 3-byte hashing
+.br
+Minimum value for
+.IR nice :
+3
+.br
+Memory usage:
+.br
+.I dict
+* 7.5 (if
+.I dict
+<= 16 MiB);
+.br
+.I dict
+* 5.5 + 64 MiB (if
+.I dict
+> 16 MiB)
+.TP
+.B hc4
+Hash Chain with 2-, 3-, and 4-byte hashing
+.br
+Minimum value for
+.IR nice :
+4
+.br
+Memory usage:
+.br
+.I dict
+* 7.5 (if
+.I dict
+<= 32 MiB);
+.br
+.I dict
+* 6.5 (if
+.I dict
+> 32 MiB)
+.TP
+.B bt2
+Binary Tree with 2-byte hashing
+.br
+Minimum value for
+.IR nice :
+2
+.br
+Memory usage:
+.I dict
+* 9.5
+.TP
+.B bt3
+Binary Tree with 2- and 3-byte hashing
+.br
+Minimum value for
+.IR nice :
+3
+.br
+Memory usage:
+.br
+.I dict
+* 11.5 (if
+.I dict
+<= 16 MiB);
+.br
+.I dict
+* 9.5 + 64 MiB (if
+.I dict
+> 16 MiB)
+.TP
+.B bt4
+Binary Tree with 2-, 3-, and 4-byte hashing
+.br
+Minimum value for
+.IR nice :
+4
+.br
+Memory usage:
+.br
+.I dict
+* 11.5 (if
+.I dict
+<= 32 MiB);
+.br
+.I dict
+* 10.5 (if
+.I dict
+> 32 MiB)
+.RE
+.TP
+.BI mode= mode
+Compression
+.I mode
+specifies the method to analyze
+the data produced by the match finder.
+Supported
+.I modes
+are
+.B fast
+and
+.BR normal .
+The default is
+.B fast
+for
+.I presets
+0\(en3 and
+.B normal
+for
+.I presets
+4\(en9.
+.IP ""
+Usually
+.B fast
+is used with Hash Chain match finders and
+.B normal
+with Binary Tree match finders.
+This is also what the
+.I presets
+do.
+.TP
+.BI nice= nice
+Specify what is considered to be a nice length for a match.
+Once a match of at least
+.I nice
+bytes is found, the algorithm stops
+looking for possibly better matches.
+.IP ""
+.I Nice
+can be 2\(en273 bytes.
+Higher values tend to give better compression ratio
+at the expense of speed.
+The default depends on the
+.IR preset .
+.TP
+.BI depth= depth
+Specify the maximum search depth in the match finder.
+The default is the special value of 0,
+which makes the compressor determine a reasonable
+.I depth
+from
+.I mf
+and
+.IR nice .
+.IP ""
+Reasonable
+.I depth
+for Hash Chains is 4\(en100 and 16\(en1000 for Binary Trees.
+Using very high values for
+.I depth
+can make the encoder extremely slow with some files.
+Avoid setting the
+.I depth
+over 1000 unless you are prepared to interrupt
+the compression in case it is taking far too long.
+.RE
+.IP ""
+When decoding raw streams
+.RB ( \-\-format=raw ),
+LZMA2 needs only the dictionary
+.IR size .
+LZMA1 needs also
+.IR lc ,
+.IR lp ,
+and
+.IR pb .
+.TP
+\fB\-\-x86\fR[\fB=\fIoptions\fR]
+.PD 0
+.TP
+\fB\-\-arm\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-armthumb\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-arm64\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-powerpc\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-ia64\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-sparc\fR[\fB=\fIoptions\fR]
+.PD
+Add a branch/call/jump (BCJ) filter to the filter chain.
+These filters can be used only as a non-last filter
+in the filter chain.
+.IP ""
+A BCJ filter converts relative addresses in
+the machine code to their absolute counterparts.
+This doesn't change the size of the data
+but it increases redundancy,
+which can help LZMA2 to produce 0\(en15\ % smaller
+.B .xz
+file.
+The BCJ filters are always reversible,
+so using a BCJ filter for wrong type of data
+doesn't cause any data loss, although it may make
+the compression ratio slightly worse.
+The BCJ filters are very fast and
+use an insignificant amount of memory.
+.IP ""
+These BCJ filters have known problems related to
+the compression ratio:
+.RS
+.IP \(bu 3
+Some types of files containing executable code
+(for example, object files, static libraries, and Linux kernel modules)
+have the addresses in the instructions filled with filler values.
+These BCJ filters will still do the address conversion,
+which will make the compression worse with these files.
+.IP \(bu 3
+If a BCJ filter is applied on an archive,
+it is possible that it makes the compression ratio
+worse than not using a BCJ filter.
+For example, if there are similar or even identical executables
+then filtering will likely make the files less similar
+and thus compression is worse.
+The contents of non-executable files in the same archive can matter too.
+In practice one has to try with and without a BCJ filter to see
+which is better in each situation.
+.RE
+.IP ""
+Different instruction sets have different alignment:
+the executable file must be aligned to a multiple of
+this value in the input data to make the filter work.
+.RS
+.RS
+.PP
+.TS
+tab(;);
+l n l
+l n l.
+Filter;Alignment;Notes
+x86;1;32-bit or 64-bit x86
+ARM;4;
+ARM-Thumb;2;
+ARM64;4;4096-byte alignment is best
+PowerPC;4;Big endian only
+IA-64;16;Itanium
+SPARC;4;
+.TE
+.RE
+.RE
+.IP ""
+Since the BCJ-filtered data is usually compressed with LZMA2,
+the compression ratio may be improved slightly if
+the LZMA2 options are set to match the
+alignment of the selected BCJ filter.
+For example, with the IA-64 filter, it's good to set
+.B pb=4
+or even
+.B pb=4,lp=4,lc=0
+with LZMA2 (2^4=16).
+The x86 filter is an exception;
+it's usually good to stick to LZMA2's default
+four-byte alignment when compressing x86 executables.
+.IP ""
+All BCJ filters support the same
+.IR options :
+.RS
+.TP
+.BI start= offset
+Specify the start
+.I offset
+that is used when converting between relative
+and absolute addresses.
+The
+.I offset
+must be a multiple of the alignment of the filter
+(see the table above).
+The default is zero.
+In practice, the default is good; specifying a custom
+.I offset
+is almost never useful.
+.RE
+.TP
+\fB\-\-delta\fR[\fB=\fIoptions\fR]
+Add the Delta filter to the filter chain.
+The Delta filter can be only used as a non-last filter
+in the filter chain.
+.IP ""
+Currently only simple byte-wise delta calculation is supported.
+It can be useful when compressing, for example, uncompressed bitmap images
+or uncompressed PCM audio.
+However, special purpose algorithms may give significantly better
+results than Delta + LZMA2.
+This is true especially with audio,
+which compresses faster and better, for example, with
+.BR flac (1).
+.IP ""
+Supported
+.IR options :
+.RS
+.TP
+.BI dist= distance
+Specify the
+.I distance
+of the delta calculation in bytes.
+.I distance
+must be 1\(en256.
+The default is 1.
+.IP ""
+For example, with
+.B dist=2
+and eight-byte input A1 B1 A2 B3 A3 B5 A4 B7, the output will be
+A1 B1 01 02 01 02 01 02.
+.RE
+.
+.SS "Other options"
+.TP
+.BR \-q ", " \-\-quiet
+Suppress warnings and notices.
+Specify this twice to suppress errors too.
+This option has no effect on the exit status.
+That is, even if a warning was suppressed,
+the exit status to indicate a warning is still used.
+.TP
+.BR \-v ", " \-\-verbose
+Be verbose.
+If standard error is connected to a terminal,
+.B xz
+will display a progress indicator.
+Specifying
+.B \-\-verbose
+twice will give even more verbose output.
+.IP ""
+The progress indicator shows the following information:
+.RS
+.IP \(bu 3
+Completion percentage is shown
+if the size of the input file is known.
+That is, the percentage cannot be shown in pipes.
+.IP \(bu 3
+Amount of compressed data produced (compressing)
+or consumed (decompressing).
+.IP \(bu 3
+Amount of uncompressed data consumed (compressing)
+or produced (decompressing).
+.IP \(bu 3
+Compression ratio, which is calculated by dividing
+the amount of compressed data processed so far by
+the amount of uncompressed data processed so far.
+.IP \(bu 3
+Compression or decompression speed.
+This is measured as the amount of uncompressed data consumed
+(compression) or produced (decompression) per second.
+It is shown after a few seconds have passed since
+.B xz
+started processing the file.
+.IP \(bu 3
+Elapsed time in the format M:SS or H:MM:SS.
+.IP \(bu 3
+Estimated remaining time is shown
+only when the size of the input file is
+known and a couple of seconds have already passed since
+.B xz
+started processing the file.
+The time is shown in a less precise format which
+never has any colons, for example, 2 min 30 s.
+.RE
+.IP ""
+When standard error is not a terminal,
+.B \-\-verbose
+will make
+.B xz
+print the filename, compressed size, uncompressed size,
+compression ratio, and possibly also the speed and elapsed time
+on a single line to standard error after compressing or
+decompressing the file.
+The speed and elapsed time are included only when
+the operation took at least a few seconds.
+If the operation didn't finish, for example, due to user interruption,
+also the completion percentage is printed
+if the size of the input file is known.
+.TP
+.BR \-Q ", " \-\-no\-warn
+Don't set the exit status to 2
+even if a condition worth a warning was detected.
+This option doesn't affect the verbosity level, thus both
+.B \-\-quiet
+and
+.B \-\-no\-warn
+have to be used to not display warnings and
+to not alter the exit status.
+.TP
+.B \-\-robot
+Print messages in a machine-parsable format.
+This is intended to ease writing frontends that want to use
+.B xz
+instead of liblzma, which may be the case with various scripts.
+The output with this option enabled is meant to be stable across
+.B xz
+releases.
+See the section
+.B "ROBOT MODE"
+for details.
+.TP
+.B \-\-info\-memory
+Display, in human-readable format, how much physical memory (RAM)
+and how many processor threads
+.B xz
+thinks the system has and the memory usage limits for compression
+and decompression, and exit successfully.
+.TP
+.BR \-h ", " \-\-help
+Display a help message describing the most commonly used options,
+and exit successfully.
+.TP
+.BR \-H ", " \-\-long\-help
+Display a help message describing all features of
+.BR xz ,
+and exit successfully
+.TP
+.BR \-V ", " \-\-version
+Display the version number of
+.B xz
+and liblzma in human readable format.
+To get machine-parsable output, specify
+.B \-\-robot
+before
+.BR \-\-version .
+.
+.SH "ROBOT MODE"
+The robot mode is activated with the
+.B \-\-robot
+option.
+It makes the output of
+.B xz
+easier to parse by other programs.
+Currently
+.B \-\-robot
+is supported only together with
+.BR \-\-version ,
+.BR \-\-info\-memory ,
+and
+.BR \-\-list .
+It will be supported for compression and
+decompression in the future.
+.
+.SS Version
+.B "xz \-\-robot \-\-version"
+prints the version number of
+.B xz
+and liblzma in the following format:
+.PP
+.BI XZ_VERSION= XYYYZZZS
+.br
+.BI LIBLZMA_VERSION= XYYYZZZS
+.TP
+.I X
+Major version.
+.TP
+.I YYY
+Minor version.
+Even numbers are stable.
+Odd numbers are alpha or beta versions.
+.TP
+.I ZZZ
+Patch level for stable releases or
+just a counter for development releases.
+.TP
+.I S
+Stability.
+0 is alpha, 1 is beta, and 2 is stable.
+.I S
+should be always 2 when
+.I YYY
+is even.
+.PP
+.I XYYYZZZS
+are the same on both lines if
+.B xz
+and liblzma are from the same XZ Utils release.
+.PP
+Examples: 4.999.9beta is
+.B 49990091
+and
+5.0.0 is
+.BR 50000002 .
+.
+.SS "Memory limit information"
+.B "xz \-\-robot \-\-info\-memory"
+prints a single line with multiple tab-separated columns:
+.IP 1. 4
+Total amount of physical memory (RAM) in bytes.
+.IP 2. 4
+Memory usage limit for compression in bytes
+.RB ( \-\-memlimit\-compress ).
+A special value of
+.B 0
+indicates the default setting
+which for single-threaded mode is the same as no limit.
+.IP 3. 4
+Memory usage limit for decompression in bytes
+.RB ( \-\-memlimit\-decompress ).
+A special value of
+.B 0
+indicates the default setting
+which for single-threaded mode is the same as no limit.
+.IP 4. 4
+Since
+.B xz
+5.3.4alpha:
+Memory usage for multi-threaded decompression in bytes
+.RB ( \-\-memlimit\-mt\-decompress ).
+This is never zero because a system-specific default value
+shown in the column 5
+is used if no limit has been specified explicitly.
+This is also never greater than the value in the column 3
+even if a larger value has been specified with
+.BR \-\-memlimit\-mt\-decompress .
+.IP 5. 4
+Since
+.B xz
+5.3.4alpha:
+A system-specific default memory usage limit
+that is used to limit the number of threads
+when compressing with an automatic number of threads
+.RB ( \-\-threads=0 )
+and no memory usage limit has been specified
+.RB ( \-\-memlimit\-compress ).
+This is also used as the default value for
+.BR \-\-memlimit\-mt\-decompress .
+.IP 6. 4
+Since
+.B xz
+5.3.4alpha:
+Number of available processor threads.
+.PP
+In the future, the output of
+.B "xz \-\-robot \-\-info\-memory"
+may have more columns, but never more than a single line.
+.
+.SS "List mode"
+.B "xz \-\-robot \-\-list"
+uses tab-separated output.
+The first column of every line has a string
+that indicates the type of the information found on that line:
+.TP
+.B name
+This is always the first line when starting to list a file.
+The second column on the line is the filename.
+.TP
+.B file
+This line contains overall information about the
+.B .xz
+file.
+This line is always printed after the
+.B name
+line.
+.TP
+.B stream
+This line type is used only when
+.B \-\-verbose
+was specified.
+There are as many
+.B stream
+lines as there are streams in the
+.B .xz
+file.
+.TP
+.B block
+This line type is used only when
+.B \-\-verbose
+was specified.
+There are as many
+.B block
+lines as there are blocks in the
+.B .xz
+file.
+The
+.B block
+lines are shown after all the
+.B stream
+lines; different line types are not interleaved.
+.TP
+.B summary
+This line type is used only when
+.B \-\-verbose
+was specified twice.
+This line is printed after all
+.B block
+lines.
+Like the
+.B file
+line, the
+.B summary
+line contains overall information about the
+.B .xz
+file.
+.TP
+.B totals
+This line is always the very last line of the list output.
+It shows the total counts and sizes.
+.PP
+The columns of the
+.B file
+lines:
+.PD 0
+.RS
+.IP 2. 4
+Number of streams in the file
+.IP 3. 4
+Total number of blocks in the stream(s)
+.IP 4. 4
+Compressed size of the file
+.IP 5. 4
+Uncompressed size of the file
+.IP 6. 4
+Compression ratio, for example,
+.BR 0.123 .
+If ratio is over 9.999, three dashes
+.RB ( \-\-\- )
+are displayed instead of the ratio.
+.IP 7. 4
+Comma-separated list of integrity check names.
+The following strings are used for the known check types:
+.BR None ,
+.BR CRC32 ,
+.BR CRC64 ,
+and
+.BR SHA\-256 .
+For unknown check types,
+.BI Unknown\- N
+is used, where
+.I N
+is the Check ID as a decimal number (one or two digits).
+.IP 8. 4
+Total size of stream padding in the file
+.RE
+.PD
+.PP
+The columns of the
+.B stream
+lines:
+.PD 0
+.RS
+.IP 2. 4
+Stream number (the first stream is 1)
+.IP 3. 4
+Number of blocks in the stream
+.IP 4. 4
+Compressed start offset
+.IP 5. 4
+Uncompressed start offset
+.IP 6. 4
+Compressed size (does not include stream padding)
+.IP 7. 4
+Uncompressed size
+.IP 8. 4
+Compression ratio
+.IP 9. 4
+Name of the integrity check
+.IP 10. 4
+Size of stream padding
+.RE
+.PD
+.PP
+The columns of the
+.B block
+lines:
+.PD 0
+.RS
+.IP 2. 4
+Number of the stream containing this block
+.IP 3. 4
+Block number relative to the beginning of the stream
+(the first block is 1)
+.IP 4. 4
+Block number relative to the beginning of the file
+.IP 5. 4
+Compressed start offset relative to the beginning of the file
+.IP 6. 4
+Uncompressed start offset relative to the beginning of the file
+.IP 7. 4
+Total compressed size of the block (includes headers)
+.IP 8. 4
+Uncompressed size
+.IP 9. 4
+Compression ratio
+.IP 10. 4
+Name of the integrity check
+.RE
+.PD
+.PP
+If
+.B \-\-verbose
+was specified twice, additional columns are included on the
+.B block
+lines.
+These are not displayed with a single
+.BR \-\-verbose ,
+because getting this information requires many seeks
+and can thus be slow:
+.PD 0
+.RS
+.IP 11. 4
+Value of the integrity check in hexadecimal
+.IP 12. 4
+Block header size
+.IP 13. 4
+Block flags:
+.B c
+indicates that compressed size is present, and
+.B u
+indicates that uncompressed size is present.
+If the flag is not set, a dash
+.RB ( \- )
+is shown instead to keep the string length fixed.
+New flags may be added to the end of the string in the future.
+.IP 14. 4
+Size of the actual compressed data in the block (this excludes
+the block header, block padding, and check fields)
+.IP 15. 4
+Amount of memory (in bytes) required to decompress
+this block with this
+.B xz
+version
+.IP 16. 4
+Filter chain.
+Note that most of the options used at compression time
+cannot be known, because only the options
+that are needed for decompression are stored in the
+.B .xz
+headers.
+.RE
+.PD
+.PP
+The columns of the
+.B summary
+lines:
+.PD 0
+.RS
+.IP 2. 4
+Amount of memory (in bytes) required to decompress
+this file with this
+.B xz
+version
+.IP 3. 4
+.B yes
+or
+.B no
+indicating if all block headers have both compressed size and
+uncompressed size stored in them
+.PP
+.I Since
+.B xz
+.I 5.1.2alpha:
+.IP 4. 4
+Minimum
+.B xz
+version required to decompress the file
+.RE
+.PD
+.PP
+The columns of the
+.B totals
+line:
+.PD 0
+.RS
+.IP 2. 4
+Number of streams
+.IP 3. 4
+Number of blocks
+.IP 4. 4
+Compressed size
+.IP 5. 4
+Uncompressed size
+.IP 6. 4
+Average compression ratio
+.IP 7. 4
+Comma-separated list of integrity check names
+that were present in the files
+.IP 8. 4
+Stream padding size
+.IP 9. 4
+Number of files.
+This is here to
+keep the order of the earlier columns the same as on
+.B file
+lines.
+.PD
+.RE
+.PP
+If
+.B \-\-verbose
+was specified twice, additional columns are included on the
+.B totals
+line:
+.PD 0
+.RS
+.IP 10. 4
+Maximum amount of memory (in bytes) required to decompress
+the files with this
+.B xz
+version
+.IP 11. 4
+.B yes
+or
+.B no
+indicating if all block headers have both compressed size and
+uncompressed size stored in them
+.PP
+.I Since
+.B xz
+.I 5.1.2alpha:
+.IP 12. 4
+Minimum
+.B xz
+version required to decompress the file
+.RE
+.PD
+.PP
+Future versions may add new line types and
+new columns can be added to the existing line types,
+but the existing columns won't be changed.
+.
+.SH "EXIT STATUS"
+.TP
+.B 0
+All is good.
+.TP
+.B 1
+An error occurred.
+.TP
+.B 2
+Something worth a warning occurred,
+but no actual errors occurred.
+.PP
+Notices (not warnings or errors) printed on standard error
+don't affect the exit status.
+.
+.SH ENVIRONMENT
+.B xz
+parses space-separated lists of options
+from the environment variables
+.B XZ_DEFAULTS
+and
+.BR XZ_OPT ,
+in this order, before parsing the options from the command line.
+Note that only options are parsed from the environment variables;
+all non-options are silently ignored.
+Parsing is done with
+.BR getopt_long (3)
+which is used also for the command line arguments.
+.TP
+.B XZ_DEFAULTS
+User-specific or system-wide default options.
+Typically this is set in a shell initialization script to enable
+.BR xz 's
+memory usage limiter by default.
+Excluding shell initialization scripts
+and similar special cases, scripts must never set or unset
+.BR XZ_DEFAULTS .
+.TP
+.B XZ_OPT
+This is for passing options to
+.B xz
+when it is not possible to set the options directly on the
+.B xz
+command line.
+This is the case when
+.B xz
+is run by a script or tool, for example, GNU
+.BR tar (1):
+.RS
+.RS
+.PP
+.nf
+.ft CW
+XZ_OPT=\-2v tar caf foo.tar.xz foo
+.ft R
+.fi
+.RE
+.RE
+.IP ""
+Scripts may use
+.BR XZ_OPT ,
+for example, to set script-specific default compression options.
+It is still recommended to allow users to override
+.B XZ_OPT
+if that is reasonable.
+For example, in
+.BR sh (1)
+scripts one may use something like this:
+.RS
+.RS
+.PP
+.nf
+.ft CW
+XZ_OPT=${XZ_OPT\-"\-7e"}
+export XZ_OPT
+.ft R
+.fi
+.RE
+.RE
+.
+.SH "LZMA UTILS COMPATIBILITY"
+The command line syntax of
+.B xz
+is practically a superset of
+.BR lzma ,
+.BR unlzma ,
+and
+.B lzcat
+as found from LZMA Utils 4.32.x.
+In most cases, it is possible to replace
+LZMA Utils with XZ Utils without breaking existing scripts.
+There are some incompatibilities though,
+which may sometimes cause problems.
+.
+.SS "Compression preset levels"
+The numbering of the compression level presets is not identical in
+.B xz
+and LZMA Utils.
+The most important difference is how dictionary sizes
+are mapped to different presets.
+Dictionary size is roughly equal to the decompressor memory usage.
+.RS
+.PP
+.TS
+tab(;);
+c c c
+c n n.
+Level;xz;LZMA Utils
+\-0;256 KiB;N/A
+\-1;1 MiB;64 KiB
+\-2;2 MiB;1 MiB
+\-3;4 MiB;512 KiB
+\-4;4 MiB;1 MiB
+\-5;8 MiB;2 MiB
+\-6;8 MiB;4 MiB
+\-7;16 MiB;8 MiB
+\-8;32 MiB;16 MiB
+\-9;64 MiB;32 MiB
+.TE
+.RE
+.PP
+The dictionary size differences affect
+the compressor memory usage too,
+but there are some other differences between
+LZMA Utils and XZ Utils, which
+make the difference even bigger:
+.RS
+.PP
+.TS
+tab(;);
+c c c
+c n n.
+Level;xz;LZMA Utils 4.32.x
+\-0;3 MiB;N/A
+\-1;9 MiB;2 MiB
+\-2;17 MiB;12 MiB
+\-3;32 MiB;12 MiB
+\-4;48 MiB;16 MiB
+\-5;94 MiB;26 MiB
+\-6;94 MiB;45 MiB
+\-7;186 MiB;83 MiB
+\-8;370 MiB;159 MiB
+\-9;674 MiB;311 MiB
+.TE
+.RE
+.PP
+The default preset level in LZMA Utils is
+.B \-7
+while in XZ Utils it is
+.BR \-6 ,
+so both use an 8 MiB dictionary by default.
+.
+.SS "Streamed vs. non-streamed .lzma files"
+The uncompressed size of the file can be stored in the
+.B .lzma
+header.
+LZMA Utils does that when compressing regular files.
+The alternative is to mark that uncompressed size is unknown
+and use end-of-payload marker to indicate
+where the decompressor should stop.
+LZMA Utils uses this method when uncompressed size isn't known,
+which is the case, for example, in pipes.
+.PP
+.B xz
+supports decompressing
+.B .lzma
+files with or without end-of-payload marker, but all
+.B .lzma
+files created by
+.B xz
+will use end-of-payload marker and have uncompressed size
+marked as unknown in the
+.B .lzma
+header.
+This may be a problem in some uncommon situations.
+For example, a
+.B .lzma
+decompressor in an embedded device might work
+only with files that have known uncompressed size.
+If you hit this problem, you need to use LZMA Utils
+or LZMA SDK to create
+.B .lzma
+files with known uncompressed size.
+.
+.SS "Unsupported .lzma files"
+The
+.B .lzma
+format allows
+.I lc
+values up to 8, and
+.I lp
+values up to 4.
+LZMA Utils can decompress files with any
+.I lc
+and
+.IR lp ,
+but always creates files with
+.B lc=3
+and
+.BR lp=0 .
+Creating files with other
+.I lc
+and
+.I lp
+is possible with
+.B xz
+and with LZMA SDK.
+.PP
+The implementation of the LZMA1 filter in liblzma
+requires that the sum of
+.I lc
+and
+.I lp
+must not exceed 4.
+Thus,
+.B .lzma
+files, which exceed this limitation, cannot be decompressed with
+.BR xz .
+.PP
+LZMA Utils creates only
+.B .lzma
+files which have a dictionary size of
+.RI "2^" n
+(a power of 2) but accepts files with any dictionary size.
+liblzma accepts only
+.B .lzma
+files which have a dictionary size of
+.RI "2^" n
+or
+.RI "2^" n " + 2^(" n "\-1)."
+This is to decrease false positives when detecting
+.B .lzma
+files.
+.PP
+These limitations shouldn't be a problem in practice,
+since practically all
+.B .lzma
+files have been compressed with settings that liblzma will accept.
+.
+.SS "Trailing garbage"
+When decompressing,
+LZMA Utils silently ignore everything after the first
+.B .lzma
+stream.
+In most situations, this is a bug.
+This also means that LZMA Utils
+don't support decompressing concatenated
+.B .lzma
+files.
+.PP
+If there is data left after the first
+.B .lzma
+stream,
+.B xz
+considers the file to be corrupt unless
+.B \-\-single\-stream
+was used.
+This may break obscure scripts which have
+assumed that trailing garbage is ignored.
+.
+.SH NOTES
+.
+.SS "Compressed output may vary"
+The exact compressed output produced from
+the same uncompressed input file
+may vary between XZ Utils versions even if
+compression options are identical.
+This is because the encoder can be improved
+(faster or better compression)
+without affecting the file format.
+The output can vary even between different
+builds of the same XZ Utils version,
+if different build options are used.
+.PP
+The above means that once
+.B \-\-rsyncable
+has been implemented,
+the resulting files won't necessarily be rsyncable
+unless both old and new files have been compressed
+with the same xz version.
+This problem can be fixed if a part of the encoder
+implementation is frozen to keep rsyncable output
+stable across xz versions.
+.
+.SS "Embedded .xz decompressors"
+Embedded
+.B .xz
+decompressor implementations like XZ Embedded don't necessarily
+support files created with integrity
+.I check
+types other than
+.B none
+and
+.BR crc32 .
+Since the default is
+.BR \-\-check=crc64 ,
+you must use
+.B \-\-check=none
+or
+.B \-\-check=crc32
+when creating files for embedded systems.
+.PP
+Outside embedded systems, all
+.B .xz
+format decompressors support all the
+.I check
+types, or at least are able to decompress
+the file without verifying the
+integrity check if the particular
+.I check
+is not supported.
+.PP
+XZ Embedded supports BCJ filters,
+but only with the default start offset.
+.
+.SH EXAMPLES
+.
+.SS Basics
+Compress the file
+.I foo
+into
+.I foo.xz
+using the default compression level
+.RB ( \-6 ),
+and remove
+.I foo
+if compression is successful:
+.RS
+.PP
+.nf
+.ft CW
+xz foo
+.ft R
+.fi
+.RE
+.PP
+Decompress
+.I bar.xz
+into
+.I bar
+and don't remove
+.I bar.xz
+even if decompression is successful:
+.RS
+.PP
+.nf
+.ft CW
+xz \-dk bar.xz
+.ft R
+.fi
+.RE
+.PP
+Create
+.I baz.tar.xz
+with the preset
+.B \-4e
+.RB ( "\-4 \-\-extreme" ),
+which is slower than the default
+.BR \-6 ,
+but needs less memory for compression and decompression (48\ MiB
+and 5\ MiB, respectively):
+.RS
+.PP
+.nf
+.ft CW
+tar cf \- baz | xz \-4e > baz.tar.xz
+.ft R
+.fi
+.RE
+.PP
+A mix of compressed and uncompressed files can be decompressed
+to standard output with a single command:
+.RS
+.PP
+.nf
+.ft CW
+xz \-dcf a.txt b.txt.xz c.txt d.txt.lzma > abcd.txt
+.ft R
+.fi
+.RE
+.
+.SS "Parallel compression of many files"
+On GNU and *BSD,
+.BR find (1)
+and
+.BR xargs (1)
+can be used to parallelize compression of many files:
+.RS
+.PP
+.nf
+.ft CW
+find . \-type f \e! \-name '*.xz' \-print0 \e
+ | xargs \-0r \-P4 \-n16 xz \-T1
+.ft R
+.fi
+.RE
+.PP
+The
+.B \-P
+option to
+.BR xargs (1)
+sets the number of parallel
+.B xz
+processes.
+The best value for the
+.B \-n
+option depends on how many files there are to be compressed.
+If there are only a couple of files,
+the value should probably be 1;
+with tens of thousands of files,
+100 or even more may be appropriate to reduce the number of
+.B xz
+processes that
+.BR xargs (1)
+will eventually create.
+.PP
+The option
+.B \-T1
+for
+.B xz
+is there to force it to single-threaded mode, because
+.BR xargs (1)
+is used to control the amount of parallelization.
+.
+.SS "Robot mode"
+Calculate how many bytes have been saved in total
+after compressing multiple files:
+.RS
+.PP
+.nf
+.ft CW
+xz \-\-robot \-\-list *.xz | awk '/^totals/{print $5\-$4}'
+.ft R
+.fi
+.RE
+.PP
+A script may want to know that it is using new enough
+.BR xz .
+The following
+.BR sh (1)
+script checks that the version number of the
+.B xz
+tool is at least 5.0.0.
+This method is compatible with old beta versions,
+which didn't support the
+.B \-\-robot
+option:
+.RS
+.PP
+.nf
+.ft CW
+if ! eval "$(xz \-\-robot \-\-version 2> /dev/null)" ||
+ [ "$XZ_VERSION" \-lt 50000002 ]; then
+ echo "Your xz is too old."
+fi
+unset XZ_VERSION LIBLZMA_VERSION
+.ft R
+.fi
+.RE
+.PP
+Set a memory usage limit for decompression using
+.BR XZ_OPT ,
+but if a limit has already been set, don't increase it:
+.RS
+.PP
+.nf
+.ft CW
+NEWLIM=$((123 << 20))\ \ # 123 MiB
+OLDLIM=$(xz \-\-robot \-\-info\-memory | cut \-f3)
+if [ $OLDLIM \-eq 0 \-o $OLDLIM \-gt $NEWLIM ]; then
+ XZ_OPT="$XZ_OPT \-\-memlimit\-decompress=$NEWLIM"
+ export XZ_OPT
+fi
+.ft R
+.fi
+.RE
+.
+.SS "Custom compressor filter chains"
+The simplest use for custom filter chains is
+customizing a LZMA2 preset.
+This can be useful,
+because the presets cover only a subset of the
+potentially useful combinations of compression settings.
+.PP
+The CompCPU columns of the tables
+from the descriptions of the options
+.BR "\-0" " ... " "\-9"
+and
+.B \-\-extreme
+are useful when customizing LZMA2 presets.
+Here are the relevant parts collected from those two tables:
+.RS
+.PP
+.TS
+tab(;);
+c c
+n n.
+Preset;CompCPU
+\-0;0
+\-1;1
+\-2;2
+\-3;3
+\-4;4
+\-5;5
+\-6;6
+\-5e;7
+\-6e;8
+.TE
+.RE
+.PP
+If you know that a file requires
+somewhat big dictionary (for example, 32\ MiB) to compress well,
+but you want to compress it quicker than
+.B "xz \-8"
+would do, a preset with a low CompCPU value (for example, 1)
+can be modified to use a bigger dictionary:
+.RS
+.PP
+.nf
+.ft CW
+xz \-\-lzma2=preset=1,dict=32MiB foo.tar
+.ft R
+.fi
+.RE
+.PP
+With certain files, the above command may be faster than
+.B "xz \-6"
+while compressing significantly better.
+However, it must be emphasized that only some files benefit from
+a big dictionary while keeping the CompCPU value low.
+The most obvious situation,
+where a big dictionary can help a lot,
+is an archive containing very similar files
+of at least a few megabytes each.
+The dictionary size has to be significantly bigger
+than any individual file to allow LZMA2 to take
+full advantage of the similarities between consecutive files.
+.PP
+If very high compressor and decompressor memory usage is fine,
+and the file being compressed is
+at least several hundred megabytes, it may be useful
+to use an even bigger dictionary than the 64 MiB that
+.B "xz \-9"
+would use:
+.RS
+.PP
+.nf
+.ft CW
+xz \-vv \-\-lzma2=dict=192MiB big_foo.tar
+.ft R
+.fi
+.RE
+.PP
+Using
+.B \-vv
+.RB ( "\-\-verbose \-\-verbose" )
+like in the above example can be useful
+to see the memory requirements
+of the compressor and decompressor.
+Remember that using a dictionary bigger than
+the size of the uncompressed file is waste of memory,
+so the above command isn't useful for small files.
+.PP
+Sometimes the compression time doesn't matter,
+but the decompressor memory usage has to be kept low, for example,
+to make it possible to decompress the file on an embedded system.
+The following command uses
+.B \-6e
+.RB ( "\-6 \-\-extreme" )
+as a base and sets the dictionary to only 64\ KiB.
+The resulting file can be decompressed with XZ Embedded
+(that's why there is
+.BR \-\-check=crc32 )
+using about 100\ KiB of memory.
+.RS
+.PP
+.nf
+.ft CW
+xz \-\-check=crc32 \-\-lzma2=preset=6e,dict=64KiB foo
+.ft R
+.fi
+.RE
+.PP
+If you want to squeeze out as many bytes as possible,
+adjusting the number of literal context bits
+.RI ( lc )
+and number of position bits
+.RI ( pb )
+can sometimes help.
+Adjusting the number of literal position bits
+.RI ( lp )
+might help too, but usually
+.I lc
+and
+.I pb
+are more important.
+For example, a source code archive contains mostly US-ASCII text,
+so something like the following might give
+slightly (like 0.1\ %) smaller file than
+.B "xz \-6e"
+(try also without
+.BR lc=4 ):
+.RS
+.PP
+.nf
+.ft CW
+xz \-\-lzma2=preset=6e,pb=0,lc=4 source_code.tar
+.ft R
+.fi
+.RE
+.PP
+Using another filter together with LZMA2 can improve
+compression with certain file types.
+For example, to compress a x86-32 or x86-64 shared library
+using the x86 BCJ filter:
+.RS
+.PP
+.nf
+.ft CW
+xz \-\-x86 \-\-lzma2 libfoo.so
+.ft R
+.fi
+.RE
+.PP
+Note that the order of the filter options is significant.
+If
+.B \-\-x86
+is specified after
+.BR \-\-lzma2 ,
+.B xz
+will give an error,
+because there cannot be any filter after LZMA2,
+and also because the x86 BCJ filter cannot be used
+as the last filter in the chain.
+.PP
+The Delta filter together with LZMA2
+can give good results with bitmap images.
+It should usually beat PNG,
+which has a few more advanced filters than simple
+delta but uses Deflate for the actual compression.
+.PP
+The image has to be saved in uncompressed format,
+for example, as uncompressed TIFF.
+The distance parameter of the Delta filter is set
+to match the number of bytes per pixel in the image.
+For example, 24-bit RGB bitmap needs
+.BR dist=3 ,
+and it is also good to pass
+.B pb=0
+to LZMA2 to accommodate the three-byte alignment:
+.RS
+.PP
+.nf
+.ft CW
+xz \-\-delta=dist=3 \-\-lzma2=pb=0 foo.tiff
+.ft R
+.fi
+.RE
+.PP
+If multiple images have been put into a single archive (for example,
+.BR .tar ),
+the Delta filter will work on that too as long as all images
+have the same number of bytes per pixel.
+.
+.SH "SEE ALSO"
+.BR xzdec (1),
+.BR xzdiff (1),
+.BR xzgrep (1),
+.BR xzless (1),
+.BR xzmore (1),
+.BR gzip (1),
+.BR bzip2 (1),
+.BR 7z (1)
+.PP
+XZ Utils: <https://tukaani.org/xz/>
+.br
+XZ Embedded: <https://tukaani.org/xz/embedded.html>
+.br
+LZMA SDK: <https://7-zip.org/sdk.html>
diff --git a/src/xz/xz_w32res.rc b/src/xz/xz_w32res.rc
new file mode 100644
index 0000000..bad3020
--- /dev/null
+++ b/src/xz/xz_w32res.rc
@@ -0,0 +1,12 @@
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#define MY_TYPE VFT_APP
+#define MY_NAME "xz"
+#define MY_SUFFIX ".exe"
+#define MY_DESC "xz data compression tool for .xz and .lzma files"
+#include "common_w32res.rc"
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-22 06:23:22 +0000