lzip/tarlz
1
0
Fork 0
Code Issues Activity

Merging upstream version 0.26.

Browse source 
Signed-off-by: Daniel Baumann <daniel@debian.org>
This commit is contained in:
Daniel Baumann 2025-02-17 21:27:57 +01:00
parent 7185f44b62
commit 180f99b04d
Signed by: daniel
GPG key ID: FBB4F0E80A80222F
44 changed files with 610 additions and 505 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
COPYING
View file

@ -1,8 +1,7 @@
GNU GENERAL PUBLIC LICENSE GNU GENERAL PUBLIC LICENSE
Version 2, June 1991 Version 2, June 1991
Copyright (C) 1989, 1991 Free Software Foundation, Inc., Copyright (C) 1989, 1991 Free Software Foundation, Inc. <http://fsf.org/>
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Everyone is permitted to copy and distribute verbatim copies Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed. of this license document, but changing it is not allowed.

24
ChangeLog
View file

@ -1,7 +1,20 @@
2024-12-07 Antonio Diaz Diaz <antonio@gnu.org>
* Version 0.26 released.
* decode.cc (decode), delete.cc (delete_members):
Return 2 if any empty lzip member is found in a multimember archive.
* create_lz.cc, decode_lz.cc:
Change 'deliver_packet' to 'deliver_packets'.
* create.cc (copy_file): Show file name if read error.
* tarlz.texi: New chapter 'Syntax of command-line arguments'.
* check.sh: Use 'cp' instead of 'cat'.
Skip time stamps out of range or not recognized by system tools.
(Reported by J Dean).
2024-01-03 Antonio Diaz Diaz <antonio@gnu.org> 2024-01-03 Antonio Diaz Diaz <antonio@gnu.org>
* Version 0.25 released. * Version 0.25 released.
* New option '--ignore-metadata. * New option '--ignore-metadata'.
* create.cc, decode.cc, decode_lz.cc: * create.cc, decode.cc, decode_lz.cc:
'#include <sys/types.h>' for major, minor, makedev on BSD systems. '#include <sys/types.h>' for major, minor, makedev on BSD systems.
* compress.cc: Reformat file diagnostics as 'PROGRAM: FILE: MESSAGE'. * compress.cc: Reformat file diagnostics as 'PROGRAM: FILE: MESSAGE'.
@ -74,7 +87,7 @@
2020-11-21 Antonio Diaz Diaz <antonio@gnu.org> 2020-11-21 Antonio Diaz Diaz <antonio@gnu.org>
* Version 0.18 released. * Version 0.18 released.
* main.cc: New option '--check-lib'. * New option '--check-lib'.
* Implement multi-threaded '-x, --extract'. * Implement multi-threaded '-x, --extract'.
* Don't #include <sys/sysmacros.h> when compiling on OS2. * Don't #include <sys/sysmacros.h> when compiling on OS2.
* delete.cc, delete_lz.cc: Use Archive_reader. * delete.cc, delete_lz.cc: Use Archive_reader.
@ -201,7 +214,7 @@
* Version 0.4 released. * Version 0.4 released.
* Add some missing #includes. * Add some missing #includes.
* main.cc: Open files in binary mode on OS2. * main.cc (main): Open files in binary mode on OS2.
2018-03-19 Antonio Diaz Diaz <antonio@gnu.org> 2018-03-19 Antonio Diaz Diaz <antonio@gnu.org>
@ -231,6 +244,5 @@
Copyright (C) 2013-2024 Antonio Diaz Diaz. Copyright (C) 2013-2024 Antonio Diaz Diaz.
This file is a collection of facts, and thus it is not copyrightable, This file is a collection of facts, and thus it is not copyrightable, but just
but just in case, you have unlimited permission to copy, distribute, and in case, you have unlimited permission to copy, distribute, and modify it.
modify it.

7
Makefile.in
View file

@ -2,8 +2,8 @@
DISTNAME = $(pkgname)-$(pkgversion) DISTNAME = $(pkgname)-$(pkgversion)
INSTALL = install INSTALL = install
INSTALL_PROGRAM = $(INSTALL) -m 755 INSTALL_PROGRAM = $(INSTALL) -m 755
INSTALL_DATA = $(INSTALL) -m 644
INSTALL_DIR = $(INSTALL) -d -m 755 INSTALL_DIR = $(INSTALL) -d -m 755
INSTALL_DATA = $(INSTALL) -m 644
SHELL = /bin/sh SHELL = /bin/sh
CAN_RUN_INSTALLINFO = $(SHELL) -c "install-info --version" > /dev/null 2>&1 CAN_RUN_INSTALLINFO = $(SHELL) -c "install-info --version" > /dev/null 2>&1
@ -31,7 +31,8 @@ main.o : main.cc
# prevent 'make' from trying to remake source files # prevent 'make' from trying to remake source files
$(VPATH)/configure $(VPATH)/Makefile.in $(VPATH)/doc/$(pkgname).texi : ; $(VPATH)/configure $(VPATH)/Makefile.in $(VPATH)/doc/$(pkgname).texi : ;
%.h %.cc : ; MAKEFLAGS += -r
.SUFFIXES :
$(objs) : Makefile $(objs) : Makefile
arg_parser.o : arg_parser.h arg_parser.o : arg_parser.h
@ -149,12 +150,12 @@ dist : doc
$(DISTNAME)/testsuite/t155.tar \ $(DISTNAME)/testsuite/t155.tar \
$(DISTNAME)/testsuite/t155_fv[1-3].tar \ $(DISTNAME)/testsuite/t155_fv[1-3].tar \
$(DISTNAME)/testsuite/eoa_blocks.tar \ $(DISTNAME)/testsuite/eoa_blocks.tar \
$(DISTNAME)/testsuite/em.lz \
$(DISTNAME)/testsuite/test.txt.lz \ $(DISTNAME)/testsuite/test.txt.lz \
$(DISTNAME)/testsuite/test.txt.tar.lz \ $(DISTNAME)/testsuite/test.txt.tar.lz \
$(DISTNAME)/testsuite/test_bad[12].txt.tar.lz \ $(DISTNAME)/testsuite/test_bad[12].txt.tar.lz \
$(DISTNAME)/testsuite/test3.tar.lz \ $(DISTNAME)/testsuite/test3.tar.lz \
$(DISTNAME)/testsuite/test3_eoa[1-5].tar.lz \ $(DISTNAME)/testsuite/test3_eoa[1-5].tar.lz \
$(DISTNAME)/testsuite/test3_em[1-6].tar.lz \
$(DISTNAME)/testsuite/test3_gh[1-6].tar.lz \ $(DISTNAME)/testsuite/test3_gh[1-6].tar.lz \
$(DISTNAME)/testsuite/test3_nn.tar.lz \ $(DISTNAME)/testsuite/test3_nn.tar.lz \
$(DISTNAME)/testsuite/test3_sm[1-4].tar.lz \ $(DISTNAME)/testsuite/test3_sm[1-4].tar.lz \

17
NEWS
View file

@ -1,14 +1,13 @@
Changes in version 0.25: Changes in version 0.26:
The new option '--ignore-metadata', which makes '-d, --diff' ignore tarlz now exits with error status 2 if any empty lzip member is found in a
differences in file permissions, owner and group IDs, and modification time, multimember compressed archive.
has been added.
'#include <sys/types.h>' for major, minor, makedev on BSD systems. Scalability of parallel compressed creation and decoding has been increased.
File diagnostics of '-z' have been reformatted as 'PROGRAM: FILE: MESSAGE'. A diagnostic message for read error has been improved.
The option '-o, --output' now creates missing intermediate directories when The chapter 'Syntax of command-line arguments' has been added to the manual.
compressing to a file.
The variable MAKEINFO has been added to configure and Makefile.in. 'make check' now skips time stamps out of range or not recognized by system
tools. (Reported by J Dean).

10
README
View file

@ -1,3 +1,5 @@
See the file INSTALL for compilation and installation instructions.
Description Description
Tarlz is a massively parallel (multi-threaded) combined implementation of Tarlz is a massively parallel (multi-threaded) combined implementation of
@ -73,8 +75,8 @@ that the format of the archive is compatible with tarlz.
The diagram below shows the correspondence between each tar member (formed The diagram below shows the correspondence between each tar member (formed
by one or two headers plus optional data) in the tar archive and each lzip by one or two headers plus optional data) in the tar archive and each lzip
member in the resulting multimember tar.lz archive, when per file member in the resulting multimember tar.lz archive, when per file compression
compression is used: is used:
tar tar
+========+======+=================+===============+========+======+========+ +========+======+=================+===============+========+======+========+
@ -87,6 +89,10 @@ tar.lz
+===============+=================================================+========+ +===============+=================================================+========+
Tarlz uses Arg_parser for command-line argument parsing:
http://www.nongnu.org/arg-parser/arg_parser.html
Copyright (C) 2013-2024 Antonio Diaz Diaz. Copyright (C) 2013-2024 Antonio Diaz Diaz.
This file is free documentation: you have unlimited permission to copy, This file is free documentation: you have unlimited permission to copy,

14
archive_reader.cc
View file

@ -19,7 +19,7 @@
#include <algorithm> #include <algorithm>
#include <cerrno> #include <cerrno>
#include <stdint.h> // for lzlib.h #include <stdint.h> // for lzlib.h
#include <unistd.h> #include <unistd.h>
#include <lzlib.h> #include <lzlib.h>
@ -116,12 +116,12 @@ int Archive_reader::read( uint8_t * const buf, const int size )
const int rd = readblock( ad.infd, buf, size ); const int rd = readblock( ad.infd, buf, size );
if( rd != size && errno ) return err( -1, rdaerr_msg, errno, rd ); if( rd != size && errno ) return err( -1, rdaerr_msg, errno, rd );
const Lzip_header & header = (*(const Lzip_header *)buf); const Lzip_header & header = (*(const Lzip_header *)buf);
const bool islz = ( rd >= min_member_size && header.check_magic() && const bool islz = rd >= min_member_size && header.check_magic() &&
header.check_version() && header.check_version() &&
isvalid_ds( header.dictionary_size() ) ); isvalid_ds( header.dictionary_size() );
const bool istar = ( rd == size && check_ustar_chksum( buf ) ); const bool istar = rd == size && check_ustar_chksum( buf );
const bool iseoa = const bool iseoa =
( !islz && !istar && rd == size && block_is_zero( buf, size ) ); !islz && !istar && rd == size && block_is_zero( buf, size );
bool maybe_lz = islz; // maybe corrupt tar.lz bool maybe_lz = islz; // maybe corrupt tar.lz
if( !islz && !istar && !iseoa ) // corrupt or invalid format if( !islz && !istar && !iseoa ) // corrupt or invalid format
{ {
@ -165,7 +165,7 @@ int Archive_reader::read( uint8_t * const buf, const int size )
if( new_pos >= old_pos && new_pos < LLONG_MAX ) if( new_pos >= old_pos && new_pos < LLONG_MAX )
return err( 2, "", 0, sz, true ); return err( 2, "", 0, sz, true );
return err( -1, "Skipping to next header failed. " return err( -1, "Skipping to next header failed. "
"Lzlib 1.8 or newer required.", 0, sz ); "Lzlib 1.8 or newer required.", 0, sz );
} }
if( rd == 0 && LZ_decompress_finished( decoder ) == 1 ) if( rd == 0 && LZ_decompress_finished( decoder ) == 1 )
{ return err( -2, end_msg, 0, sz ); } { return err( -2, end_msg, 0, sz ); }

5
archive_reader.h
View file

@ -18,7 +18,7 @@
struct Archive_descriptor struct Archive_descriptor
{ {
const std::string name; const std::string name;
const char * const namep; // printable archive name const char * const namep; // printable archive name
const int infd; const int infd;
const Lzip_index lzip_index; const Lzip_index lzip_index;
const bool seekable; const bool seekable;
@ -118,3 +118,6 @@ public:
int read( uint8_t * const buf, const int size ); int read( uint8_t * const buf, const int size );
int skip_member( const Extended & extended ); int skip_member( const Extended & extended );
}; };
const char * const empty_msg = "Empty lzip member not allowed.";

15
arg_parser.cc
View file

@ -75,19 +75,19 @@ bool Arg_parser::parse_long_option( const char * const opt, const char * const a
error_ += "' requires an argument"; error_ += "' requires an argument";
return false; return false;
} }
data.back().argument = &opt[len+3]; data.back().argument = &opt[len+3]; // argument may be empty
return true; return true;
} }
if( options[index].has_arg == yes ) if( options[index].has_arg == yes || options[index].has_arg == yme )
{ {
if( !arg || !arg[0] ) if( !arg || ( options[index].has_arg == yes && !arg[0] ) )
{ {
error_ = "option '--"; error_ += options[index].long_name; error_ = "option '--"; error_ += options[index].long_name;
error_ += "' requires an argument"; error_ += "' requires an argument";
return false; return false;
} }
++argind; data.back().argument = arg; ++argind; data.back().argument = arg; // argument may be empty
return true; return true;
} }
@ -123,15 +123,16 @@ bool Arg_parser::parse_short_option( const char * const opt, const char * const
{ {
data.back().argument = &opt[cind]; ++argind; cind = 0; data.back().argument = &opt[cind]; ++argind; cind = 0;
} }
else if( options[index].has_arg == yes ) else if( options[index].has_arg == yes || options[index].has_arg == yme )
{ {
if( !arg || !arg[0] ) if( !arg || ( options[index].has_arg == yes && !arg[0] ) )
{ {
error_ = "option requires an argument -- '"; error_ += c; error_ = "option requires an argument -- '"; error_ += c;
error_ += '\''; error_ += '\'';
return false; return false;
} }
data.back().argument = arg; ++argind; cind = 0; ++argind; cind = 0;
data.back().argument = arg; // argument may be empty
} }
} }
return true; return true;

10
arg_parser.h
View file

@ -36,14 +36,18 @@
The argument '--' terminates all options; any following arguments are The argument '--' terminates all options; any following arguments are
treated as non-option arguments, even if they begin with a hyphen. treated as non-option arguments, even if they begin with a hyphen.
The syntax for optional option arguments is '-<short_option><argument>' The syntax of options with an optional argument is
(without whitespace), or '--<long_option>=<argument>'. '-<short_option><argument>' (without whitespace), or
'--<long_option>=<argument>'.
The syntax of options with an empty argument is '-<short_option> ""',
'--<long_option> ""', or '--<long_option>=""'.
*/ */
class Arg_parser class Arg_parser
{ {
public: public:
enum Has_arg { no, yes, maybe }; enum Has_arg { no, yes, maybe, yme }; // yme = yes but maybe empty
struct Option struct Option
{ {

6
common_decode.cc
View file

@ -89,7 +89,7 @@ bool compare_prefix_dir( const char * const dir, const char * const name )
{ {
int len = 0; int len = 0;
while( dir[len] && dir[len] == name[len] ) ++len; while( dir[len] && dir[len] == name[len] ) ++len;
return ( !dir[len] && len > 0 && ( dir[len-1] == '/' || name[len] == '/' ) ); return !dir[len] && len > 0 && ( dir[len-1] == '/' || name[len] == '/' );
} }
@ -101,7 +101,7 @@ bool compare_tslash( const char * const name1, const char * const name2 )
while( *p && *p == *q ) { ++p; ++q; } while( *p && *p == *q ) { ++p; ++q; }
while( *p == '/' ) ++p; while( *p == '/' ) ++p;
while( *q == '/' ) ++q; while( *q == '/' ) ++q;
return ( !*p && !*q ); return !*p && !*q;
} }
} // end namespace } // end namespace
@ -129,7 +129,7 @@ bool format_member_name( const Extended & extended, const Tar_header header,
{ time_t z = 0; if( !gmtime_r( &z, &t ) ) // use UTC, the epoch { time_t z = 0; if( !gmtime_r( &z, &t ) ) // use UTC, the epoch
{ t.tm_year = 70; t.tm_mon = t.tm_hour = t.tm_min = 0; t.tm_mday = 1; } } { t.tm_year = 70; t.tm_mon = t.tm_hour = t.tm_min = 0; t.tm_mday = 1; } }
const Typeflag typeflag = (Typeflag)header[typeflag_o]; const Typeflag typeflag = (Typeflag)header[typeflag_o];
const bool islink = ( typeflag == tf_link || typeflag == tf_symlink ); const bool islink = typeflag == tf_link || typeflag == tf_symlink;
const char * const link_string = !islink ? "" : const char * const link_string = !islink ? "" :
( ( typeflag == tf_link ) ? " link to " : " -> " ); ( ( typeflag == tf_link ) ? " link to " : " -> " );
// print "user/group size" in a field of width 19 with 8 or more for size // print "user/group size" in a field of width 19 with 8 or more for size

9
compress.cc
View file

@ -20,7 +20,7 @@
#include <cerrno> #include <cerrno>
#include <csignal> #include <csignal>
#include <cstdio> #include <cstdio>
#include <stdint.h> // for lzlib.h #include <stdint.h> // for lzlib.h
#include <unistd.h> #include <unistd.h>
#include <utime.h> #include <utime.h>
#include <sys/stat.h> #include <sys/stat.h>
@ -134,7 +134,7 @@ bool archive_write( const uint8_t * const buf, const int size,
static bool flushed = true; // avoid flushing empty lzip members static bool flushed = true; // avoid flushing empty lzip members
if( size <= 0 && flushed ) return true; if( size <= 0 && flushed ) return true;
flushed = ( size <= 0 ); flushed = size <= 0;
enum { obuf_size = 65536 }; enum { obuf_size = 65536 };
uint8_t obuf[obuf_size]; uint8_t obuf[obuf_size];
int sz = 0; int sz = 0;
@ -150,7 +150,8 @@ bool archive_write( const uint8_t * const buf, const int size,
if( rd < 0 ) internal_error( "library error (LZ_compress_read)." ); if( rd < 0 ) internal_error( "library error (LZ_compress_read)." );
if( rd == 0 && sz >= size ) break; if( rd == 0 && sz >= size ) break;
if( writeblock( outfd, obuf, rd ) != rd ) if( writeblock( outfd, obuf, rd ) != rd )
{ show_file_error( ne_output_filename(), werr_msg, errno ); return false; } { show_file_error( ne_output_filename(), wr_err_msg, errno );
return false; }
} }
if( LZ_compress_finished( encoder ) == 1 && if( LZ_compress_finished( encoder ) == 1 &&
LZ_compress_restart_member( encoder, LLONG_MAX ) < 0 ) LZ_compress_restart_member( encoder, LLONG_MAX ) < 0 )
@ -166,7 +167,7 @@ bool tail_compress( const Cl_options & cl_opts,
if( cl_opts.solidity != solid && !archive_write( 0, 0, encoder ) ) if( cl_opts.solidity != solid && !archive_write( 0, 0, encoder ) )
return false; // flush encoder before compressing EOA blocks return false; // flush encoder before compressing EOA blocks
int size = header_size; int size = header_size;
bool zero = true; // true until non-zero data found after EOA blocks bool zero = true; // true until nonzero data found after EOA blocks
while( true ) while( true )
{ {
if( size > 0 && !archive_write( header, size, encoder ) ) if( size > 0 && !archive_write( header, size, encoder ) )

4
configure vendored
View file

@ -6,7 +6,7 @@
# to copy, distribute, and modify it. # to copy, distribute, and modify it.
pkgname=tarlz pkgname=tarlz
pkgversion=0.25 pkgversion=0.26
progname=tarlz progname=tarlz
srctrigger=doc/${pkgname}.texi srctrigger=doc/${pkgname}.texi
@ -112,7 +112,7 @@ while [ $# != 0 ] ; do
exit 1 ;; exit 1 ;;
esac esac
# Check if the option took a separate argument # Check whether the option took a separate argument
if [ "${arg2}" = yes ] ; then if [ "${arg2}" = yes ] ; then
if [ $# != 0 ] ; then args="${args} \"$1\"" ; shift if [ $# != 0 ] ; then args="${args} \"$1\"" ; shift
else echo "configure: Missing argument to '${option}'" 1>&2 else echo "configure: Missing argument to '${option}'" 1>&2

25
create.cc
View file

@ -20,14 +20,14 @@
#include <algorithm> #include <algorithm>
#include <cerrno> #include <cerrno>
#include <cstdio> #include <cstdio>
#include <stdint.h> // for lzlib.h #include <stdint.h> // for lzlib.h
#include <unistd.h> #include <unistd.h>
#include <sys/stat.h> #include <sys/stat.h>
#if !defined __FreeBSD__ && !defined __OpenBSD__ && !defined __NetBSD__ && \ #if !defined __FreeBSD__ && !defined __OpenBSD__ && !defined __NetBSD__ && \
!defined __DragonFly__ && !defined __APPLE__ && !defined __OS2__ !defined __DragonFly__ && !defined __APPLE__ && !defined __OS2__
#include <sys/sysmacros.h> // for major, minor #include <sys/sysmacros.h> // for major, minor
#else #else
#include <sys/types.h> // for major, minor #include <sys/types.h> // for major, minor
#endif #endif
#include <ftw.h> #include <ftw.h>
#include <grp.h> #include <grp.h>
@ -178,7 +178,7 @@ bool archive_write( const uint8_t * const buf, const int size )
static bool flushed = true; // avoid flushing empty lzip members static bool flushed = true; // avoid flushing empty lzip members
if( size <= 0 && flushed ) return true; if( size <= 0 && flushed ) return true;
flushed = ( size <= 0 ); flushed = size <= 0;
if( !encoder ) // uncompressed if( !encoder ) // uncompressed
return writeblock_wrapper( goutfd, buf, size ); return writeblock_wrapper( goutfd, buf, size );
enum { obuf_size = 65536 }; enum { obuf_size = 65536 };
@ -307,11 +307,12 @@ bool check_tty_out( const char * const archive_namep, const int outfd,
// infd and outfd can refer to the same file if copying to a lower file // infd and outfd can refer to the same file if copying to a lower file
// position or if source and destination blocks don't overlap. // position or if source and destination blocks don't overlap.
// max_size < 0 means no size limit. // max_size < 0 means no size limit.
bool copy_file( const int infd, const int outfd, const long long max_size ) bool copy_file( const int infd, const int outfd, const char * const filename,
const long long max_size )
{ {
const long long buffer_size = 65536; const long long buffer_size = 65536;
// remaining number of bytes to copy // remaining number of bytes to copy
long long rest = ( ( max_size >= 0 ) ? max_size : buffer_size ); long long rest = ( max_size >= 0 ) ? max_size : buffer_size;
long long copied_size = 0; long long copied_size = 0;
uint8_t * const buffer = new uint8_t[buffer_size]; uint8_t * const buffer = new uint8_t[buffer_size];
bool error = false; bool error = false;
@ -322,7 +323,7 @@ bool copy_file( const int infd, const int outfd, const long long max_size )
if( max_size >= 0 ) rest -= size; if( max_size >= 0 ) rest -= size;
const int rd = readblock( infd, buffer, size ); const int rd = readblock( infd, buffer, size );
if( rd != size && errno ) if( rd != size && errno )
{ show_error( "Error reading input file", errno ); error = true; break; } { show_file_error( filename, "Read error", errno ); error = true; break; }
if( rd > 0 ) if( rd > 0 )
{ {
if( !writeblock_wrapper( outfd, buffer, rd ) ) { error = true; break; } if( !writeblock_wrapper( outfd, buffer, rd ) ) { error = true; break; }
@ -331,7 +332,7 @@ bool copy_file( const int infd, const int outfd, const long long max_size )
if( rd < size ) break; // EOF if( rd < size ) break; // EOF
} }
delete[] buffer; delete[] buffer;
return ( !error && ( max_size < 0 || copied_size == max_size ) ); return !error && ( max_size < 0 || copied_size == max_size );
} }
@ -339,7 +340,7 @@ bool writeblock_wrapper( const int outfd, const uint8_t * const buffer,
const int size ) const int size )
{ {
if( writeblock( outfd, buffer, size ) != size ) if( writeblock( outfd, buffer, size ) != size )
{ show_file_error( archive_namep, werr_msg, errno ); return false; } { show_file_error( archive_namep, wr_err_msg, errno ); return false; }
return true; return true;
} }
@ -519,8 +520,8 @@ unsigned ustar_chksum( const Tar_header header )
bool check_ustar_chksum( const Tar_header header ) bool check_ustar_chksum( const Tar_header header )
{ return ( check_ustar_magic( header ) && { return check_ustar_magic( header ) &&
ustar_chksum( header ) == parse_octal( header + chksum_o, chksum_l ) ); } ustar_chksum( header ) == parse_octal( header + chksum_o, chksum_l ); }
bool has_lz_ext( const std::string & name ) bool has_lz_ext( const std::string & name )
@ -605,7 +606,7 @@ int concatenate( const Cl_options & cl_opts )
"Not an appendable tar.lz archive." : "Not an appendable tar.lz archive." :
"Not an appendable tar archive." ); "Not an appendable tar archive." );
close( infd ); retval = 2; break; } close( infd ); retval = 2; break; }
if( !copy_file( infd, outfd, size ) || close( infd ) != 0 ) if( !copy_file( infd, outfd, filename, size ) || close( infd ) != 0 )
{ show_file_error( filename, "Error copying archive", errno ); { show_file_error( filename, "Error copying archive", errno );
eoa_pending = false; retval = 1; break; } eoa_pending = false; retval = 1; break; }
eoa_pending = true; eoa_pending = true;

88
create_lz.cc
View file

@ -22,7 +22,7 @@
#include <cstdio> #include <cstdio>
#include <queue> #include <queue>
#include <pthread.h> #include <pthread.h>
#include <stdint.h> // for lzlib.h #include <stdint.h> // for lzlib.h
#include <unistd.h> #include <unistd.h>
#include <sys/stat.h> #include <sys/stat.h>
#include <ftw.h> #include <ftw.h>
@ -60,7 +60,7 @@ public:
~Slot_tally() { xdestroy_cond( &slot_av ); xdestroy_mutex( &mutex ); } ~Slot_tally() { xdestroy_cond( &slot_av ); xdestroy_mutex( &mutex ); }
bool all_free() { return ( num_free == num_slots ); } bool all_free() { return num_free == num_slots; }
void get_slot() // wait for a free slot void get_slot() // wait for a free slot
{ {
@ -94,8 +94,8 @@ struct Ipacket // filename, file size and headers
struct Opacket // compressed data to be written to the archive struct Opacket // compressed data to be written to the archive
{ {
const uint8_t * const data; // data == 0 means end of lzip member const uint8_t * data; // data == 0 means end of lzip member
const int size; // number of bytes in data (if any) int size; // number of bytes in data (if any)
Opacket() : data( 0 ), size( 0 ) {} Opacket() : data( 0 ), size( 0 ) {}
Opacket( uint8_t * const d, const int s ) : data( d ), size( s ) {} Opacket( uint8_t * const d, const int s ) : data( d ), size( s ) {}
@ -110,11 +110,11 @@ public:
unsigned ocheck_counter; unsigned ocheck_counter;
unsigned owait_counter; unsigned owait_counter;
private: private:
int receive_worker_id; // worker queue currently receiving packets int receive_id; // worker queue currently receiving packets
int deliver_worker_id; // worker queue currently delivering packets int deliver_id; // worker queue currently delivering packets
Slot_tally slot_tally; // limits the number of input packets Slot_tally slot_tally; // limits the number of input packets
std::vector< std::queue< const Ipacket * > > ipacket_queues; std::vector< std::queue< const Ipacket * > > ipacket_queues;
std::vector< std::queue< const Opacket * > > opacket_queues; std::vector< std::queue< Opacket > > opacket_queues;
int num_working; // number of workers still running int num_working; // number of workers still running
const int num_workers; // number of workers const int num_workers; // number of workers
const unsigned out_slots; // max output packets per queue const unsigned out_slots; // max output packets per queue
@ -132,11 +132,10 @@ public:
Packet_courier( const int workers, const int in_slots, const int oslots ) Packet_courier( const int workers, const int in_slots, const int oslots )
: icheck_counter( 0 ), iwait_counter( 0 ), : icheck_counter( 0 ), iwait_counter( 0 ),
ocheck_counter( 0 ), owait_counter( 0 ), ocheck_counter( 0 ), owait_counter( 0 ),
receive_worker_id( 0 ), deliver_worker_id( 0 ), receive_id( 0 ), deliver_id( 0 ), slot_tally( in_slots ),
slot_tally( in_slots ), ipacket_queues( workers ), ipacket_queues( workers ), opacket_queues( workers ),
opacket_queues( workers ), num_working( workers ), num_working( workers ), num_workers( workers ),
num_workers( workers ), out_slots( oslots ), slot_av( workers ), out_slots( oslots ), slot_av( workers ), eof( false )
eof( false )
{ {
xinit_mutex( &imutex ); xinit_cond( &iav_or_eof ); xinit_mutex( &imutex ); xinit_cond( &iav_or_eof );
xinit_mutex( &omutex ); xinit_cond( &oav_or_exit ); xinit_mutex( &omutex ); xinit_cond( &oav_or_exit );
@ -157,9 +156,9 @@ public:
if( !ipacket->filename.empty() ) if( !ipacket->filename.empty() )
slot_tally.get_slot(); // wait for a free slot slot_tally.get_slot(); // wait for a free slot
xlock( &imutex ); xlock( &imutex );
ipacket_queues[receive_worker_id].push( ipacket ); ipacket_queues[receive_id].push( ipacket );
if( ipacket->filename.empty() && ++receive_worker_id >= num_workers ) if( ipacket->filename.empty() && ++receive_id >= num_workers )
receive_worker_id = 0; receive_id = 0;
xbroadcast( &iav_or_eof ); xbroadcast( &iav_or_eof );
xunlock( &imutex ); xunlock( &imutex );
} }
@ -194,44 +193,41 @@ public:
} }
// collect an opacket from a worker // collect an opacket from a worker
void collect_packet( const Opacket * const opacket, const int worker_id ) void collect_packet( const Opacket & opacket, const int worker_id )
{ {
xlock( &omutex ); xlock( &omutex );
if( opacket->data ) if( opacket.data )
{ {
while( opacket_queues[worker_id].size() >= out_slots ) while( opacket_queues[worker_id].size() >= out_slots )
xwait( &slot_av[worker_id], &omutex ); xwait( &slot_av[worker_id], &omutex );
} }
opacket_queues[worker_id].push( opacket ); opacket_queues[worker_id].push( opacket );
if( worker_id == deliver_worker_id ) xsignal( &oav_or_exit ); if( worker_id == deliver_id ) xsignal( &oav_or_exit );
xunlock( &omutex ); xunlock( &omutex );
} }
/* Deliver an opacket to muxer. /* Deliver opackets to muxer.
If opacket data == 0, move to next queue and wait again. */ If opacket.data == 0, skip opacket and move to next queue. */
const Opacket * deliver_packet() void deliver_packets( std::vector< Opacket > & opacket_vector )
{ {
const Opacket * opacket = 0; opacket_vector.clear();
xlock( &omutex ); xlock( &omutex );
++ocheck_counter; ++ocheck_counter;
while( true ) do {
{ while( opacket_queues[deliver_id].empty() && num_working > 0 )
while( opacket_queues[deliver_worker_id].empty() && num_working > 0 ) { ++owait_counter; xwait( &oav_or_exit, &omutex ); }
while( !opacket_queues[deliver_id].empty() )
{ {
++owait_counter; Opacket opacket = opacket_queues[deliver_id].front();
xwait( &oav_or_exit, &omutex ); opacket_queues[deliver_id].pop();
if( opacket_queues[deliver_id].size() + 1 == out_slots )
xsignal( &slot_av[deliver_id] );
if( opacket.data ) opacket_vector.push_back( opacket );
else if( ++deliver_id >= num_workers ) deliver_id = 0;
} }
if( opacket_queues[deliver_worker_id].empty() ) break;
opacket = opacket_queues[deliver_worker_id].front();
opacket_queues[deliver_worker_id].pop();
if( opacket_queues[deliver_worker_id].size() + 1 == out_slots )
xsignal( &slot_av[deliver_worker_id] );
if( opacket->data ) break;
if( ++deliver_worker_id >= num_workers ) deliver_worker_id = 0;
delete opacket; opacket = 0;
} }
while( opacket_vector.empty() && num_working > 0 );
xunlock( &omutex ); xunlock( &omutex );
return opacket;
} }
void finish() // grouper has no more packets to send void finish() // grouper has no more packets to send
@ -371,7 +367,7 @@ void loop_encode( const uint8_t * const ibuf, const int isize,
{ {
if( opos > max_packet_size ) if( opos > max_packet_size )
internal_error( "opacket size exceeded in worker." ); internal_error( "opacket size exceeded in worker." );
courier.collect_packet( new Opacket( obuf, opos ), worker_id ); courier.collect_packet( Opacket( obuf, opos ), worker_id );
opos = 0; obuf = new( std::nothrow ) uint8_t[max_packet_size]; opos = 0; obuf = new( std::nothrow ) uint8_t[max_packet_size];
if( !obuf ) { show_error( mem_msg2 ); exit_fail_mt(); } if( !obuf ) { show_error( mem_msg2 ); exit_fail_mt(); }
if( LZ_compress_finished( encoder ) == 1 ) if( LZ_compress_finished( encoder ) == 1 )
@ -421,7 +417,7 @@ extern "C" void * cworker( void * arg )
{ {
if( !flushed ) // this lzip member is not empty if( !flushed ) // this lzip member is not empty
loop_encode( 0, 0, data, opos, courier, encoder, worker_id, true ); loop_encode( 0, 0, data, opos, courier, encoder, worker_id, true );
courier.collect_packet( new Opacket, worker_id ); // end of member token courier.collect_packet( Opacket(), worker_id ); // end of member token
flushed = true; delete ipacket; continue; flushed = true; delete ipacket; continue;
} }
@ -501,15 +497,19 @@ extern "C" void * cworker( void * arg )
*/ */
void muxer( Packet_courier & courier, const int outfd ) void muxer( Packet_courier & courier, const int outfd )
{ {
std::vector< Opacket > opacket_vector;
while( true ) while( true )
{ {
const Opacket * const opacket = courier.deliver_packet(); courier.deliver_packets( opacket_vector );
if( !opacket ) break; // queue is empty. all workers exited if( opacket_vector.empty() ) break; // queue is empty. all workers exited
if( !writeblock_wrapper( outfd, opacket->data, opacket->size ) ) for( unsigned i = 0; i < opacket_vector.size(); ++i )
exit_fail_mt(); {
delete[] opacket->data; Opacket & opacket = opacket_vector[i];
delete opacket; if( !writeblock_wrapper( outfd, opacket.data, opacket.size ) )
exit_fail_mt();
delete[] opacket.data;
}
} }
} }

14
decode.cc
View file

@ -22,15 +22,15 @@
#include <cerrno> #include <cerrno>
#include <cstdio> #include <cstdio>
#include <fcntl.h> #include <fcntl.h>
#include <stdint.h> // for lzlib.h #include <stdint.h> // for lzlib.h
#include <unistd.h> #include <unistd.h>
#include <utime.h> #include <utime.h>
#include <sys/stat.h> #include <sys/stat.h>
#if !defined __FreeBSD__ && !defined __OpenBSD__ && !defined __NetBSD__ && \ #if !defined __FreeBSD__ && !defined __OpenBSD__ && !defined __NetBSD__ && \
!defined __DragonFly__ && !defined __APPLE__ && !defined __OS2__ !defined __DragonFly__ && !defined __APPLE__ && !defined __OS2__
#include <sys/sysmacros.h> // for major, minor, makedev #include <sys/sysmacros.h> // for major, minor, makedev
#else #else
#include <sys/types.h> // for major, minor, makedev #include <sys/types.h> // for major, minor, makedev
#endif #endif
#include <lzlib.h> #include <lzlib.h>
@ -190,7 +190,7 @@ int extract_member( const Cl_options & cl_opts, Archive_reader & ar,
return skip_member( ar, extended, typeflag ); return skip_member( ar, extended, typeflag );
} }
const bool islink = ( typeflag == tf_link || typeflag == tf_symlink ); const bool islink = typeflag == tf_link || typeflag == tf_symlink;
errno = 0; errno = 0;
if( !islink && if( !islink &&
( !uid_gid_in_range( extended.get_uid(), extended.get_gid() ) || ( !uid_gid_in_range( extended.get_uid(), extended.get_gid() ) ||
@ -229,7 +229,7 @@ int extract_member( const Cl_options & cl_opts, Archive_reader & ar,
} }
const int wsize = ( rest >= bufsize ) ? bufsize : rest; const int wsize = ( rest >= bufsize ) ? bufsize : rest;
if( outfd >= 0 && writeblock( outfd, buf, wsize ) != wsize ) if( outfd >= 0 && writeblock( outfd, buf, wsize ) != wsize )
{ show_file_error( filename, werr_msg, errno ); return 1; } { show_file_error( filename, wr_err_msg, errno ); return 1; }
rest -= wsize; rest -= wsize;
} }
} }
@ -347,7 +347,7 @@ bool compare_file_type( std::string & estr, std::string & ostr,
{ {
char * const buf = new char[st.st_size+1]; char * const buf = new char[st.st_size+1];
long len = readlink( filename, buf, st.st_size ); long len = readlink( filename, buf, st.st_size );
bool e = ( len != st.st_size ); bool e = len != st.st_size;
if( !e ) if( !e )
{ {
while( len > 1 && buf[len-1] == '/' ) --len; // trailing '/' while( len > 1 && buf[len-1] == '/' ) --len; // trailing '/'
@ -412,6 +412,8 @@ int decode( const Cl_options & cl_opts )
// open archive before changing working directory // open archive before changing working directory
const Archive_descriptor ad( cl_opts.archive_name ); const Archive_descriptor ad( cl_opts.archive_name );
if( ad.infd < 0 ) return 1; if( ad.infd < 0 ) return 1;
if( ad.name.size() && ad.indexed && ad.lzip_index.multi_empty() )
{ show_file_error( ad.namep, empty_msg ); close( ad.infd ); return 2; }
const bool c_present = option_C_present( cl_opts.parser ) && const bool c_present = option_C_present( cl_opts.parser ) &&
cl_opts.program_mode != m_list; cl_opts.program_mode != m_list;

107
decode_lz.cc
View file

@ -22,15 +22,15 @@
#include <cstdio> #include <cstdio>
#include <queue> #include <queue>
#include <pthread.h> #include <pthread.h>
#include <stdint.h> // for lzlib.h #include <stdint.h> // for lzlib.h
#include <unistd.h> #include <unistd.h>
#include <utime.h> #include <utime.h>
#include <sys/stat.h> #include <sys/stat.h>
#if !defined __FreeBSD__ && !defined __OpenBSD__ && !defined __NetBSD__ && \ #if !defined __FreeBSD__ && !defined __OpenBSD__ && !defined __NetBSD__ && \
!defined __DragonFly__ && !defined __APPLE__ && !defined __OS2__ !defined __DragonFly__ && !defined __APPLE__ && !defined __OS2__
#include <sys/sysmacros.h> // for major, minor, makedev #include <sys/sysmacros.h> // for major, minor, makedev
#else #else
#include <sys/types.h> // for major, minor, makedev #include <sys/types.h> // for major, minor, makedev
#endif #endif
#include <lzlib.h> #include <lzlib.h>
@ -73,8 +73,8 @@ public:
unsigned owait_counter; unsigned owait_counter;
private: private:
long error_member_id; // first lzip member with error/misalign/eoa/eof long error_member_id; // first lzip member with error/misalign/eoa/eof
int deliver_worker_id; // worker queue currently delivering packets int deliver_id; // worker queue currently delivering packets
int master_worker_id; // worker in charge if error/misalign/eoa/eof int master_id; // worker in charge if error/misalign/eoa/eof
std::vector< std::queue< const Packet * > > opacket_queues; std::vector< std::queue< const Packet * > > opacket_queues;
int num_working; // number of workers still running int num_working; // number of workers still running
const int num_workers; // number of workers const int num_workers; // number of workers
@ -90,11 +90,10 @@ private:
public: public:
Packet_courier( const int workers, const int slots ) Packet_courier( const int workers, const int slots )
: ocheck_counter( 0 ), owait_counter( 0 ), : ocheck_counter( 0 ), owait_counter( 0 ), error_member_id( -1 ),
error_member_id( -1 ), deliver_worker_id( 0 ), master_worker_id( -1 ), deliver_id( 0 ), master_id( -1 ), opacket_queues( workers ),
opacket_queues( workers ), num_working( workers ), num_working( workers ), num_workers( workers ),
num_workers( workers ), out_slots( slots ), slot_av( workers ), out_slots( slots ), slot_av( workers ), eoa_found_( false )
eoa_found_( false )
{ {
xinit_mutex( &omutex ); xinit_cond( &oav_or_exit ); xinit_mutex( &omutex ); xinit_cond( &oav_or_exit );
for( unsigned i = 0; i < slot_av.size(); ++i ) xinit_cond( &slot_av[i] ); for( unsigned i = 0; i < slot_av.size(); ++i ) xinit_cond( &slot_av[i] );
@ -111,22 +110,22 @@ public:
bool eoa_found() const { return eoa_found_; } bool eoa_found() const { return eoa_found_; }
void report_eoa() { eoa_found_ = true; } void report_eoa() { eoa_found_ = true; }
bool mastership_granted() const { return master_worker_id >= 0; } bool mastership_granted() const { return master_id >= 0; }
bool request_mastership( const long member_id, const int worker_id ) bool request_mastership( const long member_id, const int worker_id )
{ {
xlock( &omutex ); xlock( &omutex );
if( mastership_granted() ) // already granted if( mastership_granted() ) // already granted
{ xunlock( &omutex ); return ( master_worker_id == worker_id ); } { xunlock( &omutex ); return master_id == worker_id; }
if( error_member_id < 0 || error_member_id > member_id ) if( error_member_id < 0 || error_member_id > member_id )
error_member_id = member_id; error_member_id = member_id;
while( !mastership_granted() && ( worker_id != deliver_worker_id || while( !mastership_granted() && ( worker_id != deliver_id ||
!opacket_queues[deliver_worker_id].empty() ) ) !opacket_queues[deliver_id].empty() ) )
xwait( &check_master, &omutex ); xwait( &check_master, &omutex );
if( !mastership_granted() && worker_id == deliver_worker_id && if( !mastership_granted() && worker_id == deliver_id &&
opacket_queues[deliver_worker_id].empty() ) opacket_queues[deliver_id].empty() )
{ {
master_worker_id = worker_id; // grant mastership master_id = worker_id; // grant mastership
for( int i = 0; i < num_workers; ++i ) // delete all packets for( int i = 0; i < num_workers; ++i ) // delete all packets
while( !opacket_queues[i].empty() ) while( !opacket_queues[i].empty() )
opacket_queues[i].pop(); opacket_queues[i].pop();
@ -154,46 +153,47 @@ public:
{ {
const Packet * const opacket = new Packet( member_id, msg, status, errcode ); const Packet * const opacket = new Packet( member_id, msg, status, errcode );
xlock( &omutex ); xlock( &omutex );
if( ( mastership_granted() && master_worker_id != worker_id ) || if( ( mastership_granted() && master_id != worker_id ) ||
( error_member_id >= 0 && error_member_id < opacket->member_id ) ) ( error_member_id >= 0 && error_member_id < opacket->member_id ) )
{ xunlock( &omutex ); delete opacket; return false; } // reject packet { xunlock( &omutex ); delete opacket; return false; } // reject packet
while( opacket_queues[worker_id].size() >= out_slots ) while( opacket_queues[worker_id].size() >= out_slots )
xwait( &slot_av[worker_id], &omutex ); xwait( &slot_av[worker_id], &omutex );
opacket_queues[worker_id].push( opacket ); opacket_queues[worker_id].push( opacket );
if( worker_id == deliver_worker_id ) xsignal( &oav_or_exit ); if( worker_id == deliver_id ) xsignal( &oav_or_exit );
xunlock( &omutex ); xunlock( &omutex );
return true; return true;
} }
/* Deliver a packet to muxer. /* Deliver packets to muxer.
If packet.status == Packet::member_done, move to next queue. If packet.status == Packet::member_done, move to next queue.
If packet.line.empty(), wait again (empty lzip member). */ If packet.line.empty(), wait again (empty lzip member or -q). */
const Packet * deliver_packet() void deliver_packets( std::vector< const Packet * > & opacket_vector )
{ {
const Packet * opacket = 0; opacket_vector.clear();
xlock( &omutex ); xlock( &omutex );
++ocheck_counter; ++ocheck_counter;
while( true ) do {
{ while( opacket_queues[deliver_id].empty() && num_working > 0 )
while( opacket_queues[deliver_worker_id].empty() && num_working > 0 )
{ {
++owait_counter; ++owait_counter;
if( !mastership_granted() && error_member_id >= 0 ) if( !mastership_granted() && error_member_id >= 0 )
xbroadcast( &check_master ); // mastership requested not yet granted xbroadcast( &check_master ); // mastership requested not yet granted
xwait( &oav_or_exit, &omutex ); xwait( &oav_or_exit, &omutex );
} }
if( opacket_queues[deliver_worker_id].empty() ) break; while( !opacket_queues[deliver_id].empty() )
opacket = opacket_queues[deliver_worker_id].front(); {
opacket_queues[deliver_worker_id].pop(); const Packet * opacket = opacket_queues[deliver_id].front();
if( opacket_queues[deliver_worker_id].size() + 1 == out_slots ) opacket_queues[deliver_id].pop();
xsignal( &slot_av[deliver_worker_id] ); if( opacket_queues[deliver_id].size() + 1 == out_slots )
if( opacket->status == Packet::member_done && !mastership_granted() ) xsignal( &slot_av[deliver_id] );
{ if( ++deliver_worker_id >= num_workers ) deliver_worker_id = 0; } if( opacket->status == Packet::member_done && !mastership_granted() )
if( !opacket->line.empty() ) break; { if( ++deliver_id >= num_workers ) deliver_id = 0; }
delete opacket; opacket = 0; if( !opacket->line.empty() ) opacket_vector.push_back( opacket );
else delete opacket;
}
} }
while( opacket_vector.empty() && num_working > 0 );
xunlock( &omutex ); xunlock( &omutex );
return opacket;
} }
bool finished() // all packets delivered to muxer bool finished() // all packets delivered to muxer
@ -401,7 +401,7 @@ Trival extract_member_lz( const Cl_options & cl_opts,
case tf_directory: case tf_directory:
{ {
struct stat st; struct stat st;
bool exists = ( stat( filename, &st ) == 0 ); bool exists = stat( filename, &st ) == 0;
if( exists && !S_ISDIR( st.st_mode ) ) if( exists && !S_ISDIR( st.st_mode ) )
{ exists = false; std::remove( filename ); } { exists = false; std::remove( filename ); }
if( !exists && mkdir( filename, mode ) != 0 && errno != EEXIST ) if( !exists && mkdir( filename, mode ) != 0 && errno != EEXIST )
@ -446,7 +446,7 @@ Trival extract_member_lz( const Cl_options & cl_opts,
typeflag ); typeflag );
} }
const bool islink = ( typeflag == tf_link || typeflag == tf_symlink ); const bool islink = typeflag == tf_link || typeflag == tf_symlink;
errno = 0; errno = 0;
if( !islink && if( !islink &&
( !uid_gid_in_range( extended.get_uid(), extended.get_gid() ) || ( !uid_gid_in_range( extended.get_uid(), extended.get_gid() ) ||
@ -489,7 +489,7 @@ Trival extract_member_lz( const Cl_options & cl_opts,
} }
const int wsize = ( rest >= bufsize ) ? bufsize : rest; const int wsize = ( rest >= bufsize ) ? bufsize : rest;
if( outfd >= 0 && writeblock( outfd, buf, wsize ) != wsize ) if( outfd >= 0 && writeblock( outfd, buf, wsize ) != wsize )
{ format_file_error( rbuf, filename, werr_msg, errno ); { format_file_error( rbuf, filename, wr_err_msg, errno );
return Trival( rbuf(), 0, 1 ); } return Trival( rbuf(), 0, 1 ); }
rest -= wsize; rest -= wsize;
} }
@ -672,24 +672,29 @@ done:
*/ */
void muxer( const char * const archive_namep, Packet_courier & courier ) void muxer( const char * const archive_namep, Packet_courier & courier )
{ {
std::vector< const Packet * > opacket_vector;
int retval = 0; int retval = 0;
while( retval == 0 ) while( retval == 0 )
{ {
const Packet * const opacket = courier.deliver_packet(); courier.deliver_packets( opacket_vector );
if( !opacket ) break; // queue is empty. all workers exited if( opacket_vector.empty() ) break; // queue is empty. all workers exited
switch( opacket->status ) for( unsigned i = 0; i < opacket_vector.size(); ++i )
{ {
case Packet::error1: const Packet * const opacket = opacket_vector[i];
case Packet::error2: switch( opacket->status )
show_file_error( archive_namep, opacket->line.c_str(), opacket->errcode ); {
retval = ( opacket->status == Packet::error1 ) ? 1 : 2; break; case Packet::error1:
case Packet::prefix: show_error( opacket->line.c_str() ); break; case Packet::error2:
case Packet::diag: std::fputs( opacket->line.c_str(), stderr ); break; show_file_error( archive_namep, opacket->line.c_str(), opacket->errcode );
default: if( opacket->line.size() ) retval = ( opacket->status == Packet::error1 ) ? 1 : 2; break;
{ std::fputs( opacket->line.c_str(), stdout ); std::fflush( stdout ); } case Packet::prefix: show_error( opacket->line.c_str() ); break;
case Packet::diag: std::fputs( opacket->line.c_str(), stderr ); break;
default: if( opacket->line.size() )
{ std::fputs( opacket->line.c_str(), stdout ); std::fflush( stdout ); }
}
delete opacket;
} }
delete opacket;
} }
if( retval == 0 && !courier.eoa_found() ) // no worker found EOA blocks if( retval == 0 && !courier.eoa_found() ) // no worker found EOA blocks
{ show_file_error( archive_namep, end_msg ); retval = 2; } { show_file_error( archive_namep, end_msg ); retval = 2; }

8
delete.cc
View file

@ -20,7 +20,7 @@
#include <cctype> #include <cctype>
#include <cerrno> #include <cerrno>
#include <cstdio> #include <cstdio>
#include <stdint.h> // for lzlib.h #include <stdint.h> // for lzlib.h
#include <unistd.h> #include <unistd.h>
#include <lzlib.h> #include <lzlib.h>
@ -44,7 +44,7 @@ int tail_copy( const Arg_parser & parser, const Archive_descriptor & ad,
const long long rest = ad.lzip_index.file_size() - istream_pos; const long long rest = ad.lzip_index.file_size() - istream_pos;
if( istream_pos > 0 && rest > 0 && if( istream_pos > 0 && rest > 0 &&
( !safe_seek( ad.infd, istream_pos ) || ( !safe_seek( ad.infd, istream_pos ) ||
!copy_file( ad.infd, outfd, rest ) ) ) !copy_file( ad.infd, outfd, ad.namep, rest ) ) )
{ show_file_error( ad.namep, "Error during tail copy." ); { show_file_error( ad.namep, "Error during tail copy." );
return retval ? retval : 1; } return retval ? retval : 1; }
const long long ostream_pos = lseek( outfd, 0, SEEK_CUR ); const long long ostream_pos = lseek( outfd, 0, SEEK_CUR );
@ -85,6 +85,8 @@ int delete_members( const Cl_options & cl_opts )
{ show_error( "Deleting from stdin not implemented yet." ); return 1; } { show_error( "Deleting from stdin not implemented yet." ); return 1; }
const Archive_descriptor ad( cl_opts.archive_name ); const Archive_descriptor ad( cl_opts.archive_name );
if( ad.infd < 0 ) return 1; if( ad.infd < 0 ) return 1;
if( ad.name.size() && ad.indexed && ad.lzip_index.multi_empty() )
{ show_file_error( ad.namep, empty_msg ); close( ad.infd ); return 2; }
const int outfd = open_outstream( cl_opts.archive_name, false ); const int outfd = open_outstream( cl_opts.archive_name, false );
if( outfd < 0 ) { close( ad.infd ); return 1; } if( outfd < 0 ) { close( ad.infd ); return 1; }
@ -177,7 +179,7 @@ int delete_members( const Cl_options & cl_opts )
if( istream_pos == 0 ) if( istream_pos == 0 )
{ if( !safe_seek( outfd, size ) ) { retval = 1; break; } } { if( !safe_seek( outfd, size ) ) { retval = 1; break; } }
else if( !safe_seek( ad.infd, istream_pos ) || else if( !safe_seek( ad.infd, istream_pos ) ||
!copy_file( ad.infd, outfd, size ) || !copy_file( ad.infd, outfd, ad.namep, size ) ||
!safe_seek( ad.infd, pos ) ) { retval = 1; break; } !safe_seek( ad.infd, pos ) ) { retval = 1; break; }
} }
istream_pos = pos; istream_pos = pos;

8
delete_lz.cc
View file

@ -20,7 +20,7 @@
#include <cctype> #include <cctype>
#include <cerrno> #include <cerrno>
#include <cstdio> #include <cstdio>
#include <stdint.h> // for lzlib.h #include <stdint.h> // for lzlib.h
#include <unistd.h> #include <unistd.h>
#include <lzlib.h> #include <lzlib.h>
@ -35,8 +35,7 @@
*/ */
int delete_members_lz( const Cl_options & cl_opts, int delete_members_lz( const Cl_options & cl_opts,
const Archive_descriptor & ad, const Archive_descriptor & ad,
std::vector< char > & name_pending, std::vector< char > & name_pending, const int outfd )
const int outfd )
{ {
Archive_reader_i ar( ad ); // indexed reader Archive_reader_i ar( ad ); // indexed reader
Resizable_buffer rbuf; Resizable_buffer rbuf;
@ -124,7 +123,8 @@ int delete_members_lz( const Cl_options & cl_opts,
if( istream_pos == 0 ) if( istream_pos == 0 )
{ if( !safe_seek( outfd, size ) ) { retval = 1; goto done; } } { if( !safe_seek( outfd, size ) ) { retval = 1; goto done; } }
else if( !safe_seek( ad.infd, istream_pos ) || else if( !safe_seek( ad.infd, istream_pos ) ||
!copy_file( ad.infd, outfd, size ) ) { retval = 1; goto done; } !copy_file( ad.infd, outfd, ad.namep, size ) )
{ retval = 1; goto done; }
} }
istream_pos = ad.lzip_index.mblock( i ).end(); // member end istream_pos = ad.lzip_index.mblock( i ).end(); // member end
} }

5
doc/tarlz.1
View file

@ -1,5 +1,5 @@
.\" DO NOT MODIFY THIS FILE! It was generated by help2man 1.49.2. .\" DO NOT MODIFY THIS FILE! It was generated by help2man 1.49.2.
.TH TARLZ "1" "January 2024" "tarlz 0.25" "User Commands" .TH TARLZ "1" "December 2024" "tarlz 0.26" "User Commands"
.SH NAME .SH NAME
tarlz \- creates tar archives with multimember lzip compression tarlz \- creates tar archives with multimember lzip compression
.SH SYNOPSIS .SH SYNOPSIS
@ -161,11 +161,12 @@ Report bugs to lzip\-bug@nongnu.org
Tarlz home page: http://www.nongnu.org/lzip/tarlz.html Tarlz home page: http://www.nongnu.org/lzip/tarlz.html
.SH COPYRIGHT .SH COPYRIGHT
Copyright \(co 2024 Antonio Diaz Diaz. Copyright \(co 2024 Antonio Diaz Diaz.
Using lzlib 1.14\-rc1
License GPLv2+: GNU GPL version 2 or later <http://gnu.org/licenses/gpl.html> License GPLv2+: GNU GPL version 2 or later <http://gnu.org/licenses/gpl.html>
.br .br
This is free software: you are free to change and redistribute it. This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. There is NO WARRANTY, to the extent permitted by law.
Using lzlib 1.15\-rc1
Using LZ_API_VERSION = 1015
.SH "SEE ALSO" .SH "SEE ALSO"
The full documentation for The full documentation for
.B tarlz .B tarlz

213
doc/tarlz.info
View file

@ -11,12 +11,13 @@ File: tarlz.info, Node: Top, Next: Introduction, Up: (dir)
Tarlz Manual Tarlz Manual
************ ************
This manual is for Tarlz (version 0.25, 3 January 2024). This manual is for Tarlz (version 0.26, 7 December 2024).
* Menu: * Menu:
* Introduction:: Purpose and features of tarlz * Introduction:: Purpose and features of tarlz
* Invoking tarlz:: Command-line interface * Invoking tarlz:: Command-line interface
* Argument syntax:: By convention, options start with a hyphen
* Portable character set:: POSIX portable filename character set * Portable character set:: POSIX portable filename character set
* File format:: Detailed format of the compressed archive * File format:: Detailed format of the compressed archive
* Amendments to pax format:: The reasons for the differences with pax * Amendments to pax format:: The reasons for the differences with pax
@ -92,7 +93,7 @@ in a way compatible with standard tar tools. *Note crc32::.
be used to check that the format of the archive is compatible with tarlz. be used to check that the format of the archive is compatible with tarlz.
 
File: tarlz.info, Node: Invoking tarlz, Next: Portable character set, Prev: Introduction, Up: Top File: tarlz.info, Node: Invoking tarlz, Next: Argument syntax, Prev: Introduction, Up: Top
2 Invoking tarlz 2 Invoking tarlz
**************** ****************
@ -127,7 +128,7 @@ member names in the archive or given in the command line, so that
setting is used. For example '-9 --solid --uncompressed -1' is equivalent setting is used. For example '-9 --solid --uncompressed -1' is equivalent
to '-1 --solid'. to '-1 --solid'.
tarlz supports the following operations: tarlz supports the following operations:
'--help' '--help'
Print an informative help message describing the options and exit. Print an informative help message describing the options and exit.
@ -240,7 +241,7 @@ to '-1 --solid'.
used as compressor for GNU tar by using a command like used as compressor for GNU tar by using a command like
'tar -c -Hustar foo | tarlz -z -o foo.tar.lz'. Tarlz can be used as 'tar -c -Hustar foo | tarlz -z -o foo.tar.lz'. Tarlz can be used as
compressor for zupdate (zutils) by using a command like compressor for zupdate (zutils) by using a command like
'zupdate --lz="tarlz -z" foo.tar.gz'. Note that tarlz only works 'zupdate --lz='tarlz -z' foo.tar.gz'. Note that tarlz only works
reliably on archives without global headers, or with global headers reliably on archives without global headers, or with global headers
whose content can be ignored. whose content can be ignored.
@ -249,7 +250,7 @@ to '-1 --solid'.
end-of-archive block is found, and then compresses the rest of the end-of-archive block is found, and then compresses the rest of the
archive. Unless solid compression is requested, the end-of-archive archive. Unless solid compression is requested, the end-of-archive
blocks are compressed in a lzip member separated from the preceding blocks are compressed in a lzip member separated from the preceding
members and from any non-zero garbage following the end-of-archive members and from any nonzero garbage following the end-of-archive
blocks. '--compress' implies plzip argument style, not tar style. Each blocks. '--compress' implies plzip argument style, not tar style. Each
input archive is compressed to a file with the extension '.lz' added input archive is compressed to a file with the extension '.lz' added
unless the option '--output' is used. When '--output' is used, only unless the option '--output' is used. When '--output' is used, only
@ -268,8 +269,7 @@ to '-1 --solid'.
(lzlib)Library version. (lzlib)Library version.
tarlz supports the following options: *Note Argument syntax: tarlz supports the following options: *Note Argument syntax::.
(arg_parser)Argument syntax.
'-B BYTES' '-B BYTES'
'--data-size=BYTES' '--data-size=BYTES'
@ -281,17 +281,18 @@ to '-1 --solid'.
'-C DIR' '-C DIR'
'--directory=DIR' '--directory=DIR'
Change to directory DIR. When creating, appending, comparing, or Change to directory DIR. When creating, appending, comparing, or
extracting, the position of each '-C' option in the command line is extracting, the position of each option '-C' in the command line is
significant; it changes the current working directory for the following significant; it changes the current working directory for the following
FILES until a new '-C' option appears in the command line. '--list' FILES until a new option '-C' appears in the command line. '--list'
and '--delete' ignore any '-C' options specified. DIR is relative to and '--delete' ignore any option '-C' specified. DIR is relative to
the then current working directory, perhaps changed by a previous '-C' the then current working directory, perhaps changed by a previous
option. option '-C'.
Note that a process can only have one current working directory (CWD). Note that a process can only have one current working directory (CWD).
Therefore multi-threading can't be used to create or decode an archive Therefore multi-threading can't be used to create or decode an archive
if a '-C' option appears after a (relative) file name in the command if an option '-C' appears after a (relative) file name in the command
line. (All file names are made relative when decoding). line. (All file names are made relative by removing leading slashes
when decoding).
'-f ARCHIVE' '-f ARCHIVE'
'--file=ARCHIVE' '--file=ARCHIVE'
@ -307,10 +308,10 @@ to '-1 --solid'.
'-n N' '-n N'
'--threads=N' '--threads=N'
Set the number of (de)compression threads, overriding the system's Set the number of (de)compression threads, overriding the system's
default. Valid values range from 0 to "as many as your system can default. Valid values range from 0 to as many as your system can
support". A value of 0 disables threads entirely. If this option is support. A value of 0 disables threads entirely. If this option is not
not used, tarlz tries to detect the number of processors in the system used, tarlz tries to detect the number of processors in the system and
and use it as default value. 'tarlz --help' shows the system's default use it as default value. 'tarlz --help' shows the system's default
value. See the note about multi-threading in the option '-C' above. value. See the note about multi-threading in the option '-C' above.
Note that the number of usable threads is limited during compression to Note that the number of usable threads is limited during compression to
@ -347,6 +348,7 @@ to '-1 --solid'.
reducing the amount of memory required for decompression. reducing the amount of memory required for decompression.
Level Dictionary size Match length limit Level Dictionary size Match length limit
--------------------------------------------
-0 64 KiB 16 bytes -0 64 KiB 16 bytes
-1 1 MiB 5 bytes -1 1 MiB 5 bytes
-2 1.5 MiB 6 bytes -2 1.5 MiB 6 bytes
@ -496,9 +498,52 @@ indicate a corrupt or invalid input file, 3 for an internal consistency
error (e.g., bug) which caused tarlz to panic. error (e.g., bug) which caused tarlz to panic.
 
File: tarlz.info, Node: Portable character set, Next: File format, Prev: Invoking tarlz, Up: Top File: tarlz.info, Node: Argument syntax, Next: Portable character set, Prev: Invoking tarlz, Up: Top
3 POSIX portable filename character set 3 Syntax of command-line arguments
**********************************
POSIX recommends these conventions for command-line arguments.
* A command-line argument is an option if it begins with a hyphen ('-').
* Option names are single alphanumeric characters.
* Certain options require an argument.
* An option and its argument may or may not appear as separate tokens.
(In other words, the whitespace separating them is optional). Thus,
'-o foo' and '-ofoo' are equivalent.
* One or more options without arguments, followed by at most one option
that takes an argument, may follow a hyphen in a single token. Thus,
'-abc' is equivalent to '-a -b -c'.
* Options typically precede other non-option arguments.
* The argument '--' terminates all options; any following arguments are
treated as non-option arguments, even if they begin with a hyphen.
* A token consisting of a single hyphen character is interpreted as an
ordinary non-option argument. By convention, it is used to specify
standard input, standard output, or a file named '-'.
GNU adds "long options" to these conventions:
* A long option consists of two hyphens ('--') followed by a name made
of alphanumeric characters and hyphens. Option names are typically one
to three words long, with hyphens to separate words. Abbreviations can
be used for the long option names as long as the abbreviations are
unique.
* A long option and its argument may or may not appear as separate
tokens. In the latter case they must be separated by an equal sign '='.
Thus, '--foo bar' and '--foo=bar' are equivalent.

File: tarlz.info, Node: Portable character set, Next: File format, Prev: Argument syntax, Up: Top
4 POSIX portable filename character set
*************************************** ***************************************
The set of characters from which portable file names are constructed. The set of characters from which portable file names are constructed.
@ -516,7 +561,7 @@ names use only the portable character set without spaces added.
 
File: tarlz.info, Node: File format, Next: Amendments to pax format, Prev: Portable character set, Up: Top File: tarlz.info, Node: File format, Next: Amendments to pax format, Prev: Portable character set, Up: Top
4 File format 5 File format
************* *************
In the diagram below, a box like this: In the diagram below, a box like this:
@ -534,10 +579,10 @@ In the diagram below, a box like this:
represents a variable number of bytes or a fixed but large number of represents a variable number of bytes or a fixed but large number of
bytes (for example 512). bytes (for example 512).
A tar.lz file consists of one or more lzip members (compressed data sets).
A tar.lz file consists of one or more lzip members (compressed data The members simply appear one after another in the file, with no additional
sets). The members simply appear one after another in the file, with no information before, between, or after them. Empty members (data size = 0)
additional information before, between, or after them. are not allowed in multimember files.
Each lzip member contains one or more tar members in a simplified POSIX Each lzip member contains one or more tar members in a simplified POSIX
pax interchange format. The only pax typeflag value supported by tarlz (in pax interchange format. The only pax typeflag value supported by tarlz (in
@ -570,7 +615,7 @@ binary zeros, interpreted as an end-of-archive indicator. These EOA blocks
are either compressed in a separate lzip member or compressed along with the are either compressed in a separate lzip member or compressed along with the
tar members contained in the last lzip member. For a compressed archive to tar members contained in the last lzip member. For a compressed archive to
be recognized by tarlz as appendable, the last lzip member must contain be recognized by tarlz as appendable, the last lzip member must contain
between 512 and 32256 zeros alone (without any non-zero bytes). between 512 and 32256 zeros alone (without any nonzero bytes).
The diagram below shows the correspondence between each tar member The diagram below shows the correspondence between each tar member
(formed by one or two headers plus optional data) in the tar archive and (formed by one or two headers plus optional data) in the tar archive and
@ -587,15 +632,14 @@ tar.lz
| member | member | member | | member | member | member |
+===============+=================================================+========+ +===============+=================================================+========+
5.1 Pax header block
4.1 Pax header block
==================== ====================
The pax header block is identical to the ustar header block described below The pax header block is identical to the ustar header block described below
except that the typeflag has the value 'x' (extended). The field 'size' is except that the typeflag has the value 'x' (extended). The field 'size' is
the size of the extended header data in bytes. Most other fields in the pax the size of the extended header data in bytes. Most other fields in the pax
header block are zeroed on archive creation to prevent trouble if the header block are zeroed on archive creation to prevent trouble if the
archive is read by an ustar tool, and are ignored by tarlz on archive archive is read by a ustar tool, and are ignored by tarlz on archive
extraction. *Note flawed-compat::. extraction. *Note flawed-compat::.
Tarlz limits the size of the pax extended header data so that the whole Tarlz limits the size of the pax extended header data so that the whole
@ -682,14 +726,14 @@ space, equal-sign, and newline.
At verbosity level 1 or higher tarlz prints a diagnostic for each unknown At verbosity level 1 or higher tarlz prints a diagnostic for each unknown
extended header keyword found in an archive, once per keyword. extended header keyword found in an archive, once per keyword.
5.2 Ustar header block
4.2 Ustar header block
====================== ======================
The ustar header block has a length of 512 bytes and is structured as shown The ustar header block has a length of 512 bytes and is structured as shown
in the following table. All lengths and offsets are in decimal. in the following table. All lengths and offsets are in decimal:
Field Name Offset Length (in bytes) Field Name Offset Length (in bytes)
---------------------------------------
name 0 100 name 0 100
mode 100 8 mode 100 8
uid 108 8 uid 108 8
@ -810,7 +854,7 @@ longer than standard ustar by not requiring a terminating null character.
 
File: tarlz.info, Node: Amendments to pax format, Next: Program design, Prev: File format, Up: Top File: tarlz.info, Node: Amendments to pax format, Next: Program design, Prev: File format, Up: Top
5 The reasons for the differences with pax 6 The reasons for the differences with pax
****************************************** ******************************************
Tarlz creates safe archives that allow the reliable detection of invalid or Tarlz creates safe archives that allow the reliable detection of invalid or
@ -821,8 +865,7 @@ achieve this goal and avoid some other flaws in the pax format, tarlz makes
some changes to the variant of the pax format that it uses. This chapter some changes to the variant of the pax format that it uses. This chapter
describes these changes and the concrete reasons to implement them. describes these changes and the concrete reasons to implement them.
6.1 Add a CRC of the extended records
5.1 Add a CRC of the extended records
===================================== =====================================
The POSIX pax format has a serious flaw. The metadata stored in pax extended The POSIX pax format has a serious flaw. The metadata stored in pax extended
@ -849,8 +892,7 @@ place.
Redundancy Check (CRC) in a way compatible with standard tar tools. *Note Redundancy Check (CRC) in a way compatible with standard tar tools. *Note
key_crc32::. key_crc32::.
6.2 Remove flawed backward compatibility
5.2 Remove flawed backward compatibility
======================================== ========================================
In order to allow the extraction of pax archives by a tar utility conforming In order to allow the extraction of pax archives by a tar utility conforming
@ -878,13 +920,12 @@ violations during parallel extraction.
If an extended header is required for any reason (for example a file If an extended header is required for any reason (for example a file
size of 8 GiB or larger, or a link name longer than 100 bytes), tarlz also size of 8 GiB or larger, or a link name longer than 100 bytes), tarlz also
moves the file name to the extended records to prevent an ustar tool from moves the file name to the extended records to prevent a ustar tool from
trying to extract the file or link. This also makes easier during parallel trying to extract the file or link. This also makes easier during parallel
decoding the detection of a tar member split between two lzip members at decoding the detection of a tar member split between two lzip members at
the boundary between the extended header and the ustar header. the boundary between the extended header and the ustar header.
6.3 As simple as possible (but not simpler)
5.3 As simple as possible (but not simpler)
=========================================== ===========================================
The tarlz format is mainly ustar. Extended pax headers are used only when The tarlz format is mainly ustar. Extended pax headers are used only when
@ -899,8 +940,7 @@ corruption.
ignored. Some operations may not behave as expected if the archive contains ignored. Some operations may not behave as expected if the archive contains
global headers. global headers.
6.4 Improve reproducibility
5.4 Improve reproducibility
=========================== ===========================
Pax includes by default the process ID of the pax process in the ustar name Pax includes by default the process ID of the pax process in the ustar name
@ -912,8 +952,7 @@ extended records, making it easier to produce reproducible archives.
ten; '99<97_bytes>' or '100<97_bytes>'. Tarlz minimizes the length of the ten; '99<97_bytes>' or '100<97_bytes>'. Tarlz minimizes the length of the
record and always produces a length of x-1 in these cases. record and always produces a length of x-1 in these cases.
6.5 No data in hard links
5.5 No data in hard links
========================= =========================
Tarlz does not allow data in hard link members. The data (if any) must be in Tarlz does not allow data in hard link members. The data (if any) must be in
@ -922,8 +961,7 @@ the names of a file are stored as hard links, the type of the file is lost.
Not allowing data in hard links also prevents invalid actions like Not allowing data in hard links also prevents invalid actions like
extracting file data for a hard link to a symbolic link or to a directory. extracting file data for a hard link to a symbolic link or to a directory.
6.6 Avoid misconversions to/from UTF-8
5.6 Avoid misconversions to/from UTF-8
====================================== ======================================
There is no portable way to tell what charset a text string is coded into. There is no portable way to tell what charset a text string is coded into.
@ -935,7 +973,7 @@ be adjusted with a command-line option in the future.
 
File: tarlz.info, Node: Program design, Next: Multi-threaded decoding, Prev: Amendments to pax format, Up: Top File: tarlz.info, Node: Program design, Next: Multi-threaded decoding, Prev: Amendments to pax format, Up: Top
6 Internal structure of tarlz 7 Internal structure of tarlz
***************************** *****************************
The parts of tarlz related to sequential processing of the archive are more The parts of tarlz related to sequential processing of the archive are more
@ -947,7 +985,7 @@ processing.
creation is somewhat similar to that of plzip with the added complication creation is somewhat similar to that of plzip with the added complication
of the solidity levels. *Note Program design: (plzip)Program design. A of the solidity levels. *Note Program design: (plzip)Program design. A
grouper thread and several worker threads are created, acting the main grouper thread and several worker threads are created, acting the main
thread as muxer (multiplexer) thread. A "packet courier" takes care of data thread as muxer (multiplexer) thread. A 'packet courier' takes care of data
transfers among threads and limits the maximum number of data blocks transfers among threads and limits the maximum number of data blocks
(packets) being processed simultaneously. (packets) being processed simultaneously.
@ -993,8 +1031,8 @@ the archive.
As misaligned tar.lz archives can't be decoded in parallel, and the As misaligned tar.lz archives can't be decoded in parallel, and the
misalignment can't be detected until after decoding has started, a misalignment can't be detected until after decoding has started, a
"mastership request" mechanism has been designed that allows the decoding to 'mastership request' mechanism has been designed that allows the decoding to
continue instead of signalling an error. continue instead of exiting with an error.
During parallel decoding, if a worker finds a misalignment, it requests During parallel decoding, if a worker finds a misalignment, it requests
mastership to decode the rest of the archive. When mastership is requested, mastership to decode the rest of the archive. When mastership is requested,
@ -1015,15 +1053,15 @@ error be avoided.
 
File: tarlz.info, Node: Multi-threaded decoding, Next: Minimum archive sizes, Prev: Program design, Up: Top File: tarlz.info, Node: Multi-threaded decoding, Next: Minimum archive sizes, Prev: Program design, Up: Top
7 Limitations of parallel tar decoding 8 Limitations of parallel tar decoding
************************************** **************************************
Safely decoding an arbitrary tar archive in parallel is only possible if one Safely decoding a tar archive in parallel is only possible if one decodes
decodes the headers sequentially first. For example, if a tar archive the headers sequentially first. For example, if a tar archive containing
containing another tar archive is decoded starting from some position other another tar archive is decoded starting from some position other than the
than the beginning, there is no way to know if the first header found there beginning, there is no way to know if the first header found there belongs
belongs to the outer tar archive or to the inner tar archive. Tar is a to the outer tar archive or to the inner tar archive. Tar is a format
format inherently serial; it was designed for tapes. inherently serial; it was designed for tapes.
The pax format is even more serial than the ustar format. Two headers The pax format is even more serial than the ustar format. Two headers
need to be decoded sequentially for each file. The extended header may even need to be decoded sequentially for each file. The extended header may even
@ -1071,8 +1109,7 @@ the tar member data because it only decodes the part of each lzip member
corresponding to the tar member header. This is another reason why the tar corresponding to the tar member header. This is another reason why the tar
headers must provide their own integrity checking. headers must provide their own integrity checking.
8.1 Limitations of multi-threaded extraction
7.1 Limitations of multi-threaded extraction
============================================ ============================================
Multi-threaded extraction may produce different output than single-threaded Multi-threaded extraction may produce different output than single-threaded
@ -1102,7 +1139,7 @@ links to.
 
File: tarlz.info, Node: Minimum archive sizes, Next: Examples, Prev: Multi-threaded decoding, Up: Top File: tarlz.info, Node: Minimum archive sizes, Next: Examples, Prev: Multi-threaded decoding, Up: Top
8 Minimum archive sizes required for multi-threaded block compression 9 Minimum archive sizes required for multi-threaded block compression
********************************************************************* *********************************************************************
When creating or appending to a compressed archive using multi-threaded When creating or appending to a compressed archive using multi-threaded
@ -1140,21 +1177,19 @@ Level
 
File: tarlz.info, Node: Examples, Next: Problems, Prev: Minimum archive sizes, Up: Top File: tarlz.info, Node: Examples, Next: Problems, Prev: Minimum archive sizes, Up: Top
9 A small tutorial with examples 10 A small tutorial with examples
******************************** *********************************
Example 1: Create a multimember compressed archive 'archive.tar.lz' Example 1: Create a multimember compressed archive 'archive.tar.lz'
containing files 'a', 'b' and 'c'. containing files 'a', 'b' and 'c'.
tarlz -cf archive.tar.lz a b c tarlz -cf archive.tar.lz a b c
Example 2: Append files 'd' and 'e' to the multimember compressed archive Example 2: Append files 'd' and 'e' to the multimember compressed archive
'archive.tar.lz'. 'archive.tar.lz'.
tarlz -rf archive.tar.lz d e tarlz -rf archive.tar.lz d e
Example 3: Create a solidly compressed appendable archive 'archive.tar.lz' Example 3: Create a solidly compressed appendable archive 'archive.tar.lz'
containing files 'a', 'b' and 'c'. Then append files 'd' and 'e' to the containing files 'a', 'b' and 'c'. Then append files 'd' and 'e' to the
archive. archive.
@ -1162,7 +1197,6 @@ archive.
tarlz --asolid -cf archive.tar.lz a b c tarlz --asolid -cf archive.tar.lz a b c
tarlz --asolid -rf archive.tar.lz d e tarlz --asolid -rf archive.tar.lz d e
Example 4: Create a compressed appendable archive containing directories Example 4: Create a compressed appendable archive containing directories
'dir1', 'dir2' and 'dir3' with a separate lzip member per directory. Then 'dir1', 'dir2' and 'dir3' with a separate lzip member per directory. Then
append files 'a', 'b', 'c', 'd' and 'e' to the archive, all of them append files 'a', 'b', 'c', 'd' and 'e' to the archive, all of them
@ -1172,31 +1206,26 @@ contains 5 lzip members (including the end-of-archive member).
tarlz --dsolid -cf archive.tar.lz dir1 dir2 dir3 tarlz --dsolid -cf archive.tar.lz dir1 dir2 dir3
tarlz --asolid -rf archive.tar.lz a b c d e tarlz --asolid -rf archive.tar.lz a b c d e
Example 5: Create a solidly compressed archive 'archive.tar.lz' containing Example 5: Create a solidly compressed archive 'archive.tar.lz' containing
files 'a', 'b' and 'c'. Note that no more files can be later appended to files 'a', 'b' and 'c'. Note that no more files can be later appended to
the archive. the archive.
tarlz --solid -cf archive.tar.lz a b c tarlz --solid -cf archive.tar.lz a b c
Example 6: Extract all files from archive 'archive.tar.lz'. Example 6: Extract all files from archive 'archive.tar.lz'.
tarlz -xf archive.tar.lz tarlz -xf archive.tar.lz
Example 7: Extract files 'a' and 'c', and the whole tree under directory Example 7: Extract files 'a' and 'c', and the whole tree under directory
'dir1' from archive 'archive.tar.lz'. 'dir1' from archive 'archive.tar.lz'.
tarlz -xf archive.tar.lz a c dir1 tarlz -xf archive.tar.lz a c dir1
Example 8: Copy the contents of directory 'sourcedir' to the directory Example 8: Copy the contents of directory 'sourcedir' to the directory
'destdir'. 'destdir'.
tarlz -C sourcedir --uncompressed -cf - . | tarlz -C destdir -xf - tarlz -C sourcedir --uncompressed -cf - . | tarlz -C destdir -xf -
Example 9: Compress the existing POSIX archive 'archive.tar' and write the Example 9: Compress the existing POSIX archive 'archive.tar' and write the
output to 'archive.tar.lz'. Compress each member individually for maximum output to 'archive.tar.lz'. Compress each member individually for maximum
availability. (If one member in the compressed archive gets damaged, the availability. (If one member in the compressed archive gets damaged, the
@ -1204,13 +1233,11 @@ other members can still be extracted).
tarlz -z --no-solid archive.tar tarlz -z --no-solid archive.tar
Example 10: Compress the archive 'archive.tar' and write the output to Example 10: Compress the archive 'archive.tar' and write the output to
'foo.tar.lz'. 'foo.tar.lz'.
tarlz -z -o foo.tar.lz archive.tar tarlz -z -o foo.tar.lz archive.tar
Example 11: Concatenate and compress two archives 'archive1.tar' and Example 11: Concatenate and compress two archives 'archive1.tar' and
'archive2.tar', and write the output to 'foo.tar.lz'. 'archive2.tar', and write the output to 'foo.tar.lz'.
@ -1219,7 +1246,7 @@ Example 11: Concatenate and compress two archives 'archive1.tar' and
 
File: tarlz.info, Node: Problems, Next: Concept index, Prev: Examples, Up: Top File: tarlz.info, Node: Problems, Next: Concept index, Prev: Examples, Up: Top
10 Reporting bugs 11 Reporting bugs
***************** *****************
There are probably bugs in tarlz. There are certainly errors and omissions There are probably bugs in tarlz. There are certainly errors and omissions
@ -1241,6 +1268,7 @@ Concept index
* Menu: * Menu:
* Amendments to pax format: Amendments to pax format. (line 6) * Amendments to pax format: Amendments to pax format. (line 6)
* argument syntax: Argument syntax. (line 6)
* bugs: Problems. (line 6) * bugs: Problems. (line 6)
* examples: Examples. (line 6) * examples: Examples. (line 6)
* file format: File format. (line 6) * file format: File format. (line 6)
@ -1259,25 +1287,26 @@ Concept index
 
Tag Table: Tag Table:
Node: Top216 Node: Top216
Node: Introduction1207 Node: Introduction1281
Node: Invoking tarlz4032 Node: Invoking tarlz4106
Ref: --data-size13076 Ref: --data-size13109
Ref: --bsolid17512 Ref: --bsolid17626
Node: Portable character set23425 Node: Argument syntax23539
Node: File format24068 Node: Portable character set25314
Ref: key_crc3231050 Node: File format25958
Ref: ustar-uid-gid34315 Ref: key_crc3233001
Ref: ustar-mtime35122 Ref: ustar-uid-gid36305
Node: Amendments to pax format37125 Ref: ustar-mtime37112
Ref: crc3237834 Node: Amendments to pax format39115
Ref: flawed-compat39146 Ref: crc3239823
Node: Program design43228 Ref: flawed-compat41134
Node: Multi-threaded decoding47153 Node: Program design45211
Ref: mt-extraction50434 Node: Multi-threaded decoding49138
Node: Minimum archive sizes51740 Ref: mt-extraction52407
Node: Examples53867 Node: Minimum archive sizes53713
Node: Problems56234 Node: Examples55840
Node: Concept index56789 Node: Problems58199
Node: Concept index58754
 
End Tag Table End Tag Table

220
doc/tarlz.texi
View file

@ -6,8 +6,8 @@
@finalout @finalout
@c %**end of header @c %**end of header
@set UPDATED 3 January 2024 @set UPDATED 7 December 2024
@set VERSION 0.25 @set VERSION 0.26
@dircategory Archiving @dircategory Archiving
@direntry @direntry
@ -38,6 +38,7 @@ This manual is for Tarlz (version @value{VERSION}, @value{UPDATED}).
@menu @menu
* Introduction:: Purpose and features of tarlz * Introduction:: Purpose and features of tarlz
* Invoking tarlz:: Command-line interface * Invoking tarlz:: Command-line interface
* Argument syntax:: By convention, options start with a hyphen
* Portable character set:: POSIX portable filename character set * Portable character set:: POSIX portable filename character set
* File format:: Detailed format of the compressed archive * File format:: Detailed format of the compressed archive
* Amendments to pax format:: The reasons for the differences with pax * Amendments to pax format:: The reasons for the differences with pax
@ -150,21 +151,22 @@ Tarlz does not use absolute file names nor file names above the current
working directory (perhaps changed by option @option{-C}). On archive creation working directory (perhaps changed by option @option{-C}). On archive creation
or appending tarlz archives the files specified, but removes from member or appending tarlz archives the files specified, but removes from member
names any leading and trailing slashes and any file name prefixes containing names any leading and trailing slashes and any file name prefixes containing
a @samp{..} component. On extraction, leading and trailing slashes are also a @file{..} component. On extraction, leading and trailing slashes are also
removed from member names, and archive members containing a @samp{..} removed from member names, and archive members containing a @file{..}
component in the file name are skipped. Tarlz does not follow symbolic links component in the file name are skipped. Tarlz does not follow symbolic links
during extraction; not even symbolic links replacing intermediate during extraction; not even symbolic links replacing intermediate
directories. directories.
On extraction and listing, tarlz removes leading @samp{./} strings from On extraction and listing, tarlz removes leading @file{./} strings from
member names in the archive or given in the command line, so that member names in the archive or given in the command line, so that
@w{@samp{tarlz -xf foo ./bar baz}} extracts members @samp{bar} and @w{@samp{tarlz -xf foo ./bar baz}} extracts members @file{bar} and
@samp{./baz} from archive @samp{foo}. @file{./baz} from archive @file{foo}.
If several compression levels or @option{--*solid} options are given, the last If several compression levels or @option{--*solid} options are given, the last
setting is used. For example @w{@option{-9 --solid --uncompressed -1}} is setting is used. For example @w{@option{-9 --solid --uncompressed -1}} is
equivalent to @w{@option{-1 --solid}}. equivalent to @w{@option{-1 --solid}}.
@noindent
tarlz supports the following operations: tarlz supports the following operations:
@table @code @table @code
@ -277,7 +279,7 @@ standard output (unless the option @option{--output} is used). Tarlz can be
used as compressor for GNU tar by using a command like used as compressor for GNU tar by using a command like
@w{@samp{tar -c -Hustar foo | tarlz -z -o foo.tar.lz}}. Tarlz can be used as @w{@samp{tar -c -Hustar foo | tarlz -z -o foo.tar.lz}}. Tarlz can be used as
compressor for zupdate (zutils) by using a command like compressor for zupdate (zutils) by using a command like
@w{@samp{zupdate --lz="tarlz -z" foo.tar.gz}}. Note that tarlz only works @w{@samp{zupdate --lz='tarlz -z' foo.tar.gz}}. Note that tarlz only works
reliably on archives without global headers, or with global headers whose reliably on archives without global headers, or with global headers whose
content can be ignored. content can be ignored.
@ -285,10 +287,10 @@ The compression is reversible, including any garbage present after the
end-of-archive blocks. Tarlz stops parsing after the first end-of-archive end-of-archive blocks. Tarlz stops parsing after the first end-of-archive
block is found, and then compresses the rest of the archive. Unless solid block is found, and then compresses the rest of the archive. Unless solid
compression is requested, the end-of-archive blocks are compressed in a lzip compression is requested, the end-of-archive blocks are compressed in a lzip
member separated from the preceding members and from any non-zero garbage member separated from the preceding members and from any nonzero garbage
following the end-of-archive blocks. @option{--compress} implies plzip following the end-of-archive blocks. @option{--compress} implies plzip
argument style, not tar style. Each input archive is compressed to a file argument style, not tar style. Each input archive is compressed to a file
with the extension @samp{.lz} added unless the option @option{--output} is with the extension @file{.lz} added unless the option @option{--output} is
used. When @option{--output} is used, only one input archive can be specified. used. When @option{--output} is used, only one input archive can be specified.
@option{-f} can't be used with @option{--compress}. @option{-f} can't be used with @option{--compress}.
@ -308,11 +310,8 @@ and the value of LZ_API_VERSION (if defined).
@end table @end table
tarlz supports the following @noindent
@uref{http://www.nongnu.org/arg-parser/manual/arg_parser_manual.html#Argument-syntax,,options}: tarlz supports the following options: @xref{Argument syntax}.
@ifnothtml
@xref{Argument syntax,,,arg_parser}.
@end ifnothtml
@table @code @table @code
@anchor{--data-size} @anchor{--data-size}
@ -326,17 +325,17 @@ defaults to @w{1 MiB}. @xref{Minimum archive sizes}.
@item -C @var{dir} @item -C @var{dir}
@itemx --directory=@var{dir} @itemx --directory=@var{dir}
Change to directory @var{dir}. When creating, appending, comparing, or Change to directory @var{dir}. When creating, appending, comparing, or
extracting, the position of each @option{-C} option in the command line is extracting, the position of each option @option{-C} in the command line is
significant; it changes the current working directory for the following significant; it changes the current working directory for the following
@var{files} until a new @option{-C} option appears in the command line. @var{files} until a new option @option{-C} appears in the command line.
@option{--list} and @option{--delete} ignore any @option{-C} options @option{--list} and @option{--delete} ignore any option @option{-C}
specified. @var{dir} is relative to the then current working directory, specified. @var{dir} is relative to the then current working directory,
perhaps changed by a previous @option{-C} option. perhaps changed by a previous option @option{-C}.
Note that a process can only have one current working directory (CWD). Note that a process can only have one current working directory (CWD).
Therefore multi-threading can't be used to create or decode an archive if a Therefore multi-threading can't be used to create or decode an archive if an
@option{-C} option appears after a (relative) file name in the command line. option @option{-C} appears after a (relative) file name in the command line.
(All file names are made relative when decoding). (All file names are made relative by removing leading slashes when decoding).
@item -f @var{archive} @item -f @var{archive}
@itemx --file=@var{archive} @itemx --file=@var{archive}
@ -351,7 +350,7 @@ Archive or compare the files they point to instead of the links themselves.
@item -n @var{n} @item -n @var{n}
@itemx --threads=@var{n} @itemx --threads=@var{n}
Set the number of (de)compression threads, overriding the system's default. Set the number of (de)compression threads, overriding the system's default.
Valid values range from 0 to "as many as your system can support". A value Valid values range from 0 to as many as your system can support. A value
of 0 disables threads entirely. If this option is not used, tarlz tries to of 0 disables threads entirely. If this option is not used, tarlz tries to
detect the number of processors in the system and use it as default value. detect the number of processors in the system and use it as default value.
@w{@samp{tarlz --help}} shows the system's default value. See the note about @w{@samp{tarlz --help}} shows the system's default value. See the note about
@ -391,7 +390,7 @@ tarlz also minimizes the dictionary size of the lzip members it creates,
reducing the amount of memory required for decompression. reducing the amount of memory required for decompression.
@multitable {Level} {Dictionary size} {Match length limit} @multitable {Level} {Dictionary size} {Match length limit}
@item Level @tab Dictionary size @tab Match length limit @headitem Level @tab Dictionary size @tab Match length limit
@item -0 @tab 64 KiB @tab 16 bytes @item -0 @tab 64 KiB @tab 16 bytes
@item -1 @tab 1 MiB @tab 5 bytes @item -1 @tab 1 MiB @tab 5 bytes
@item -2 @tab 1.5 MiB @tab 6 bytes @item -2 @tab 1.5 MiB @tab 6 bytes
@ -410,7 +409,7 @@ uncompressed tar archive instead. With @option{--append}, don't compress the
new members appended to the tar archive. Compressed members can't be new members appended to the tar archive. Compressed members can't be
appended to an uncompressed archive, nor vice versa. @option{--uncompressed} appended to an uncompressed archive, nor vice versa. @option{--uncompressed}
can be omitted if it can be deduced from the archive name. (An uncompressed can be omitted if it can be deduced from the archive name. (An uncompressed
archive name lacks a @samp{.lz} or @samp{.tlz} extension). archive name lacks a @file{.lz} or @file{.tlz} extension).
@item --asolid @item --asolid
When creating or appending to a compressed archive, use appendable solid When creating or appending to a compressed archive, use appendable solid
@ -466,11 +465,11 @@ If @var{group} is not a valid group name, it is decoded as a decimal numeric
group ID. group ID.
@item --exclude=@var{pattern} @item --exclude=@var{pattern}
Exclude files matching a shell pattern like @samp{*.o}. A file is considered Exclude files matching a shell pattern like @file{*.o}. A file is considered
to match if any component of the file name matches. For example, @samp{*.o} to match if any component of the file name matches. For example, @file{*.o}
matches @samp{foo.o}, @samp{foo.o/bar} and @samp{foo/bar.o}. If matches @file{foo.o}, @file{foo.o/bar} and @file{foo/bar.o}. If
@var{pattern} contains a @samp{/}, it matches a corresponding @samp{/} in @var{pattern} contains a @samp{/}, it matches a corresponding @samp{/} in
the file name. For example, @samp{foo/*.o} matches @samp{foo/bar.o}. the file name. For example, @file{foo/*.o} matches @file{foo/bar.o}.
Multiple @option{--exclude} options can be specified. Multiple @option{--exclude} options can be specified.
@item --ignore-ids @item --ignore-ids
@ -545,6 +544,53 @@ etc), 2 to indicate a corrupt or invalid input file, 3 for an internal
consistency error (e.g., bug) which caused tarlz to panic. consistency error (e.g., bug) which caused tarlz to panic.
@node Argument syntax
@chapter Syntax of command-line arguments
@cindex argument syntax
POSIX recommends these conventions for command-line arguments.
@itemize @bullet
@item A command-line argument is an option if it begins with a hyphen
(@samp{-}).
@item Option names are single alphanumeric characters.
@item Certain options require an argument.
@item An option and its argument may or may not appear as separate tokens.
(In other words, the whitespace separating them is optional).
Thus, @w{@option{-o foo}} and @option{-ofoo} are equivalent.
@item One or more options without arguments, followed by at most one option
that takes an argument, may follow a hyphen in a single token.
Thus, @option{-abc} is equivalent to @w{@option{-a -b -c}}.
@item Options typically precede other non-option arguments.
@item The argument @samp{--} terminates all options; any following arguments
are treated as non-option arguments, even if they begin with a hyphen.
@item A token consisting of a single hyphen character is interpreted as an
ordinary non-option argument. By convention, it is used to specify standard
input, standard output, or a file named @samp{-}.
@end itemize
@noindent
GNU adds @dfn{long options} to these conventions:
@itemize @bullet
@item A long option consists of two hyphens (@samp{--}) followed by a name
made of alphanumeric characters and hyphens. Option names are typically one
to three words long, with hyphens to separate words. Abbreviations can be
used for the long option names as long as the abbreviations are unique.
@item A long option and its argument may or may not appear as separate
tokens. In the latter case they must be separated by an equal sign @samp{=}.
Thus, @w{@option{--foo bar}} and @option{--foo=bar} are equivalent.
@end itemize
@node Portable character set @node Portable character set
@chapter POSIX portable filename character set @chapter POSIX portable filename character set
@cindex portable character set @cindex portable character set
@ -587,10 +633,11 @@ represents one byte; a box like this:
represents a variable number of bytes or a fixed but large number of represents a variable number of bytes or a fixed but large number of
bytes (for example 512). bytes (for example 512).
@sp 1 @noindent
A tar.lz file consists of one or more lzip members (compressed data sets). A tar.lz file consists of one or more lzip members (compressed data sets).
The members simply appear one after another in the file, with no additional The members simply appear one after another in the file, with no additional
information before, between, or after them. information before, between, or after them. Empty members (data size = 0)
are not allowed in multimember files.
Each lzip member contains one or more tar members in a simplified POSIX pax Each lzip member contains one or more tar members in a simplified POSIX pax
interchange format. The only pax typeflag value supported by tarlz (in interchange format. The only pax typeflag value supported by tarlz (in
@ -628,7 +675,7 @@ binary zeros, interpreted as an end-of-archive indicator. These EOA blocks
are either compressed in a separate lzip member or compressed along with the are either compressed in a separate lzip member or compressed along with the
tar members contained in the last lzip member. For a compressed archive to tar members contained in the last lzip member. For a compressed archive to
be recognized by tarlz as appendable, the last lzip member must contain be recognized by tarlz as appendable, the last lzip member must contain
between 512 and 32256 zeros alone (without any non-zero bytes). between 512 and 32256 zeros alone (without any nonzero bytes).
The diagram below shows the correspondence between each tar member (formed The diagram below shows the correspondence between each tar member (formed
by one or two headers plus optional data) in the tar archive and each by one or two headers plus optional data) in the tar archive and each
@ -652,25 +699,24 @@ tar.lz
@end verbatim @end verbatim
@ignore @ignore
When @option{--permissive} is used, the following violations of the When @option{--permissive} is used, the following violations of the archive
archive format are allowed:@* format are allowed:@*
If several extended headers precede an ustar header, only the last If several extended headers precede a ustar header, only the last extended
extended header takes effect. The other extended headers are ignored. header takes effect. The other extended headers are ignored. Similarly, if
Similarly, if several records with the same keyword appear in the same several records with the same keyword appear in the same block of extended
block of extended records, only the last record for the repeated keyword records, only the last record for the repeated keyword takes effect. The
takes effect. The other records for the repeated keyword are ignored.@* other records for the repeated keyword are ignored.@*
A global header inserted between an extended header and an ustar header.@* A global header inserted between an extended header and a ustar header.@*
An extended header just before the end-of-archive blocks. An extended header just before the end-of-archive blocks.
@end ignore @end ignore
@sp 1
@section Pax header block @section Pax header block
The pax header block is identical to the ustar header block described below The pax header block is identical to the ustar header block described below
except that the typeflag has the value @samp{x} (extended). The field except that the typeflag has the value @samp{x} (extended). The field
@samp{size} is the size of the extended header data in bytes. Most other @samp{size} is the size of the extended header data in bytes. Most other
fields in the pax header block are zeroed on archive creation to prevent fields in the pax header block are zeroed on archive creation to prevent
trouble if the archive is read by an ustar tool, and are ignored by tarlz on trouble if the archive is read by a ustar tool, and are ignored by tarlz on
archive extraction. @xref{flawed-compat}. archive extraction. @xref{flawed-compat}.
Tarlz limits the size of the pax extended header data so that the whole Tarlz limits the size of the pax extended header data so that the whole
@ -756,14 +802,13 @@ swapping of two bytes.
At verbosity level 1 or higher tarlz prints a diagnostic for each unknown At verbosity level 1 or higher tarlz prints a diagnostic for each unknown
extended header keyword found in an archive, once per keyword. extended header keyword found in an archive, once per keyword.
@sp 1
@section Ustar header block @section Ustar header block
The ustar header block has a length of 512 bytes and is structured as The ustar header block has a length of 512 bytes and is structured as
shown in the following table. All lengths and offsets are in decimal. shown in the following table. All lengths and offsets are in decimal:
@multitable {Field Name} {Offset} {Length (in bytes)} @multitable {Field Name} {Offset} {Length (in bytes)}
@item Field Name @tab Offset @tab Length (in bytes) @headitem Field Name @tab Offset @tab Length (in bytes)
@item name @tab 0 @tab 100 @item name @tab 0 @tab 100
@item mode @tab 100 @tab 8 @item mode @tab 100 @tab 8
@item uid @tab 108 @tab 8 @item uid @tab 108 @tab 8
@ -901,7 +946,6 @@ In order to achieve this goal and avoid some other flaws in the pax format,
tarlz makes some changes to the variant of the pax format that it uses. This tarlz makes some changes to the variant of the pax format that it uses. This
chapter describes these changes and the concrete reasons to implement them. chapter describes these changes and the concrete reasons to implement them.
@sp 1
@anchor{crc32} @anchor{crc32}
@section Add a CRC of the extended records @section Add a CRC of the extended records
@ -928,7 +972,6 @@ Because of the above, tarlz protects the extended records with a Cyclic
Redundancy Check (CRC) in a way compatible with standard tar tools. Redundancy Check (CRC) in a way compatible with standard tar tools.
@xref{key_crc32}. @xref{key_crc32}.
@sp 1
@anchor{flawed-compat} @anchor{flawed-compat}
@section Remove flawed backward compatibility @section Remove flawed backward compatibility
@ -958,12 +1001,11 @@ extraction.
If an extended header is required for any reason (for example a file size of If an extended header is required for any reason (for example a file size of
@w{8 GiB} or larger, or a link name longer than 100 bytes), tarlz also moves @w{8 GiB} or larger, or a link name longer than 100 bytes), tarlz also moves
the file name to the extended records to prevent an ustar tool from trying the file name to the extended records to prevent a ustar tool from trying to
to extract the file or link. This also makes easier during parallel decoding extract the file or link. This also makes easier during parallel decoding
the detection of a tar member split between two lzip members at the boundary the detection of a tar member split between two lzip members at the boundary
between the extended header and the ustar header. between the extended header and the ustar header.
@sp 1
@section As simple as possible (but not simpler) @section As simple as possible (but not simpler)
The tarlz format is mainly ustar. Extended pax headers are used only when The tarlz format is mainly ustar. Extended pax headers are used only when
@ -978,7 +1020,6 @@ Global pax headers are tolerated, but not supported; they are parsed and
ignored. Some operations may not behave as expected if the archive contains ignored. Some operations may not behave as expected if the archive contains
global headers. global headers.
@sp 1
@section Improve reproducibility @section Improve reproducibility
Pax includes by default the process ID of the pax process in the ustar name Pax includes by default the process ID of the pax process in the ustar name
@ -990,7 +1031,6 @@ Pax allows an extended record to have length x-1 or x if x is a power of
ten; @samp{99<97_bytes>} or @samp{100<97_bytes>}. Tarlz minimizes the length ten; @samp{99<97_bytes>} or @samp{100<97_bytes>}. Tarlz minimizes the length
of the record and always produces a length of x-1 in these cases. of the record and always produces a length of x-1 in these cases.
@sp 1
@section No data in hard links @section No data in hard links
Tarlz does not allow data in hard link members. The data (if any) must be in Tarlz does not allow data in hard link members. The data (if any) must be in
@ -999,7 +1039,6 @@ the names of a file are stored as hard links, the type of the file is lost.
Not allowing data in hard links also prevents invalid actions like Not allowing data in hard links also prevents invalid actions like
extracting file data for a hard link to a symbolic link or to a directory. extracting file data for a hard link to a symbolic link or to a directory.
@sp 1
@section Avoid misconversions to/from UTF-8 @section Avoid misconversions to/from UTF-8
There is no portable way to tell what charset a text string is coded into. There is no portable way to tell what charset a text string is coded into.
@ -1025,7 +1064,7 @@ added complication of the solidity levels.
@xref{Program design,,,plzip}. @xref{Program design,,,plzip}.
@end ifnothtml @end ifnothtml
A grouper thread and several worker threads are created, acting the main A grouper thread and several worker threads are created, acting the main
thread as muxer (multiplexer) thread. A "packet courier" takes care of data thread as muxer (multiplexer) thread. A 'packet courier' takes care of data
transfers among threads and limits the maximum number of data blocks transfers among threads and limits the maximum number of data blocks
(packets) being processed simultaneously. (packets) being processed simultaneously.
@ -1074,8 +1113,8 @@ access files in the file system either to read them (diff) or write them
As misaligned tar.lz archives can't be decoded in parallel, and the As misaligned tar.lz archives can't be decoded in parallel, and the
misalignment can't be detected until after decoding has started, a misalignment can't be detected until after decoding has started, a
"mastership request" mechanism has been designed that allows the decoding to 'mastership request' mechanism has been designed that allows the decoding to
continue instead of signalling an error. continue instead of exiting with an error.
During parallel decoding, if a worker finds a misalignment, it requests During parallel decoding, if a worker finds a misalignment, it requests
mastership to decode the rest of the archive. When mastership is requested, mastership to decode the rest of the archive. When mastership is requested,
@ -1098,12 +1137,12 @@ error be avoided.
@chapter Limitations of parallel tar decoding @chapter Limitations of parallel tar decoding
@cindex parallel tar decoding @cindex parallel tar decoding
Safely decoding an arbitrary tar archive in parallel is only possible if one Safely decoding a tar archive in parallel is only possible if one decodes
decodes the headers sequentially first. For example, if a tar archive the headers sequentially first. For example, if a tar archive containing
containing another tar archive is decoded starting from some position other another tar archive is decoded starting from some position other than the
than the beginning, there is no way to know if the first header found there beginning, there is no way to know if the first header found there belongs
belongs to the outer tar archive or to the inner tar archive. Tar is a to the outer tar archive or to the inner tar archive. Tar is a format
format inherently serial; it was designed for tapes. inherently serial; it was designed for tapes.
The pax format is even more serial than the ustar format. Two headers need The pax format is even more serial than the ustar format. Two headers need
to be decoded sequentially for each file. The extended header may even need to be decoded sequentially for each file. The extended header may even need
@ -1153,7 +1192,6 @@ the tar member data because it only decodes the part of each lzip member
corresponding to the tar member header. This is another reason why the tar corresponding to the tar member header. This is another reason why the tar
headers must provide their own integrity checking. headers must provide their own integrity checking.
@sp 1
@anchor{mt-extraction} @anchor{mt-extraction}
@section Limitations of multi-threaded extraction @section Limitations of multi-threaded extraction
@ -1225,40 +1263,37 @@ data size for each level:
@cindex examples @cindex examples
@noindent @noindent
Example 1: Create a multimember compressed archive @samp{archive.tar.lz} Example 1: Create a multimember compressed archive @file{archive.tar.lz}
containing files @samp{a}, @samp{b} and @samp{c}. containing files @file{a}, @file{b} and @file{c}.
@example @example
tarlz -cf archive.tar.lz a b c tarlz -cf archive.tar.lz a b c
@end example @end example
@sp 1
@noindent @noindent
Example 2: Append files @samp{d} and @samp{e} to the multimember compressed Example 2: Append files @file{d} and @file{e} to the multimember compressed
archive @samp{archive.tar.lz}. archive @file{archive.tar.lz}.
@example @example
tarlz -rf archive.tar.lz d e tarlz -rf archive.tar.lz d e
@end example @end example
@sp 1
@noindent @noindent
Example 3: Create a solidly compressed appendable archive Example 3: Create a solidly compressed appendable archive
@samp{archive.tar.lz} containing files @samp{a}, @samp{b} and @samp{c}. @file{archive.tar.lz} containing files @file{a}, @file{b} and @file{c}.
Then append files @samp{d} and @samp{e} to the archive. Then append files @file{d} and @file{e} to the archive.
@example @example
tarlz --asolid -cf archive.tar.lz a b c tarlz --asolid -cf archive.tar.lz a b c
tarlz --asolid -rf archive.tar.lz d e tarlz --asolid -rf archive.tar.lz d e
@end example @end example
@sp 1
@noindent @noindent
Example 4: Create a compressed appendable archive containing directories Example 4: Create a compressed appendable archive containing directories
@samp{dir1}, @samp{dir2} and @samp{dir3} with a separate lzip member per @file{dir1}, @file{dir2} and @file{dir3} with a separate lzip member per
directory. Then append files @samp{a}, @samp{b}, @samp{c}, @samp{d} and directory. Then append files @file{a}, @file{b}, @file{c}, @file{d} and
@samp{e} to the archive, all of them contained in a single lzip member. @file{e} to the archive, all of them contained in a single lzip member.
The resulting archive @samp{archive.tar.lz} contains 5 lzip members The resulting archive @file{archive.tar.lz} contains 5 lzip members
(including the end-of-archive member). (including the end-of-archive member).
@example @example
@ -1266,46 +1301,41 @@ tarlz --dsolid -cf archive.tar.lz dir1 dir2 dir3
tarlz --asolid -rf archive.tar.lz a b c d e tarlz --asolid -rf archive.tar.lz a b c d e
@end example @end example
@sp 1
@noindent @noindent
Example 5: Create a solidly compressed archive @samp{archive.tar.lz} Example 5: Create a solidly compressed archive @file{archive.tar.lz}
containing files @samp{a}, @samp{b} and @samp{c}. Note that no more containing files @file{a}, @file{b} and @file{c}. Note that no more
files can be later appended to the archive. files can be later appended to the archive.
@example @example
tarlz --solid -cf archive.tar.lz a b c tarlz --solid -cf archive.tar.lz a b c
@end example @end example
@sp 1
@noindent @noindent
Example 6: Extract all files from archive @samp{archive.tar.lz}. Example 6: Extract all files from archive @file{archive.tar.lz}.
@example @example
tarlz -xf archive.tar.lz tarlz -xf archive.tar.lz
@end example @end example
@sp 1
@noindent @noindent
Example 7: Extract files @samp{a} and @samp{c}, and the whole tree under Example 7: Extract files @file{a} and @file{c}, and the whole tree under
directory @samp{dir1} from archive @samp{archive.tar.lz}. directory @file{dir1} from archive @file{archive.tar.lz}.
@example @example
tarlz -xf archive.tar.lz a c dir1 tarlz -xf archive.tar.lz a c dir1
@end example @end example
@sp 1
@noindent @noindent
Example 8: Copy the contents of directory @samp{sourcedir} to the directory Example 8: Copy the contents of directory @file{sourcedir} to the directory
@samp{destdir}. @file{destdir}.
@example @example
tarlz -C sourcedir --uncompressed -cf - . | tarlz -C destdir -xf - tarlz -C sourcedir --uncompressed -cf - . | tarlz -C destdir -xf -
@end example @end example
@sp 1
@noindent @noindent
Example 9: Compress the existing POSIX archive @samp{archive.tar} and write Example 9: Compress the existing POSIX archive @file{archive.tar} and write
the output to @samp{archive.tar.lz}. Compress each member individually for the output to @file{archive.tar.lz}. Compress each member individually for
maximum availability. (If one member in the compressed archive gets damaged, maximum availability. (If one member in the compressed archive gets damaged,
the other members can still be extracted). the other members can still be extracted).
@ -1313,19 +1343,17 @@ the other members can still be extracted).
tarlz -z --no-solid archive.tar tarlz -z --no-solid archive.tar
@end example @end example
@sp 1
@noindent @noindent
Example 10: Compress the archive @samp{archive.tar} and write the output to Example 10: Compress the archive @file{archive.tar} and write the output to
@samp{foo.tar.lz}. @file{foo.tar.lz}.
@example @example
tarlz -z -o foo.tar.lz archive.tar tarlz -z -o foo.tar.lz archive.tar
@end example @end example
@sp 1
@noindent @noindent
Example 11: Concatenate and compress two archives @samp{archive1.tar} and Example 11: Concatenate and compress two archives @file{archive1.tar} and
@samp{archive2.tar}, and write the output to @samp{foo.tar.lz}. @file{archive2.tar}, and write the output to @file{foo.tar.lz}.
@example @example
tarlz -A archive1.tar archive2.tar | tarlz -z -o foo.tar.lz tarlz -A archive1.tar archive2.tar | tarlz -z -o foo.tar.lz

17
lzip_index.cc
View file

@ -49,9 +49,8 @@ const char * bad_version( const unsigned version )
bool Lzip_index::check_header( const Lzip_header & header, const bool first ) bool Lzip_index::check_header( const Lzip_header & header, const bool first )
{ {
if( !header.check_magic() ) if( header.check_magic() ) { if( first ) good_magic_ = true; }
{ error_ = bad_magic_msg; retval_ = 2; if( first ) bad_magic_ = true; else { error_ = bad_magic_msg; retval_ = 2; return false; }
return false; }
if( !header.check_version() ) if( !header.check_version() )
{ error_ = bad_version( header.version() ); retval_ = 2; return false; } { error_ = bad_version( header.version() ); retval_ = 2; return false; }
if( !isvalid_ds( header.dictionary_size() ) ) if( !isvalid_ds( header.dictionary_size() ) )
@ -146,20 +145,20 @@ bool Lzip_index::skip_trailing_data( const int fd, unsigned long long & pos )
Lzip_index::Lzip_index( const int infd ) Lzip_index::Lzip_index( const int infd )
: insize( lseek( infd, 0, SEEK_END ) ), retval_( 0 ), dictionary_size_( 0 ), : insize( lseek( infd, 0, SEEK_END ) ), retval_( 0 ), dictionary_size_( 0 ),
bad_magic_( false ) good_magic_( false )
{ {
if( insize < 0 ) if( insize < 0 )
{ set_errno_error( "Input file is not seekable: " ); return; } { set_errno_error( "Input file is not seekable: " ); return; }
Lzip_header header;
if( insize >= header.size &&
( !read_header( infd, header, 0 ) ||
!check_header( header, true ) ) ) return;
if( insize < min_member_size ) if( insize < min_member_size )
{ error_ = "Input file is too short."; retval_ = 2; return; } { error_ = "Input file is truncated."; retval_ = 2; return; }
if( insize > INT64_MAX ) if( insize > INT64_MAX )
{ error_ = "Input file is too long (2^63 bytes or more)."; { error_ = "Input file is too long (2^63 bytes or more).";
retval_ = 2; return; } retval_ = 2; return; }
Lzip_header header;
if( !read_header( infd, header, 0 ) ||
!check_header( header, true ) ) return;
unsigned long long pos = insize; // always points to a header or to EOF unsigned long long pos = insize; // always points to a header or to EOF
while( pos >= min_member_size ) while( pos >= min_member_size )
{ {

12
lzip_index.h
View file

@ -55,7 +55,7 @@ class Lzip_index
const long long insize; const long long insize;
int retval_; int retval_;
unsigned dictionary_size_; // largest dictionary size in the file unsigned dictionary_size_; // largest dictionary size in the file
bool bad_magic_; // bad magic in first header bool good_magic_; // good magic in first header
bool check_header( const Lzip_header & header, const bool first ); bool check_header( const Lzip_header & header, const bool first );
void set_errno_error( const char * const msg ); void set_errno_error( const char * const msg );
@ -70,7 +70,15 @@ public:
const std::string & error() const { return error_; } const std::string & error() const { return error_; }
int retval() const { return retval_; } int retval() const { return retval_; }
unsigned dictionary_size() const { return dictionary_size_; } unsigned dictionary_size() const { return dictionary_size_; }
bool bad_magic() const { return bad_magic_; } bool good_magic() const { return good_magic_; }
bool multi_empty() const // multimember file with empty member(s)
{
if( member_vector.size() > 1 )
for( unsigned long i = 0; i < member_vector.size(); ++i )
if( member_vector[i].dblock.size() == 0 ) return true;
return false;
}
long long udata_size() const long long udata_size() const
{ if( member_vector.empty() ) return 0; { if( member_vector.empty() ) return 0;

46
main.cc
View file

@ -25,12 +25,13 @@
#include <cctype> #include <cctype>
#include <cerrno> #include <cerrno>
#include <climits> // CHAR_BIT
#include <cstdarg> #include <cstdarg>
#include <cstdio> #include <cstdio>
#include <ctime> #include <ctime>
#include <fcntl.h> #include <fcntl.h>
#include <pthread.h> // for pthread_t #include <pthread.h> // for pthread_t
#include <stdint.h> // for lzlib.h #include <stdint.h> // for lzlib.h
#include <unistd.h> #include <unistd.h>
#include <sys/stat.h> #include <sys/stat.h>
#include <grp.h> #include <grp.h>
@ -140,14 +141,28 @@ void show_help( const long num_online )
} }
void show_lzlib_version()
{
std::printf( "Using lzlib %s\n", LZ_version() );
#if !defined LZ_API_VERSION
std::fputs( "LZ_API_VERSION is not defined.\n", stdout );
#elif LZ_API_VERSION >= 1012
std::printf( "Using LZ_API_VERSION = %u\n", LZ_api_version() );
#else
std::printf( "Compiled with LZ_API_VERSION = %u. "
"Using an unknown LZ_API_VERSION\n", LZ_API_VERSION );
#endif
}
void show_version() void show_version()
{ {
std::printf( "%s %s\n", program_name, PROGVERSION ); std::printf( "%s %s\n", program_name, PROGVERSION );
std::printf( "Copyright (C) %s Antonio Diaz Diaz.\n", program_year ); std::printf( "Copyright (C) %s Antonio Diaz Diaz.\n", program_year );
std::printf( "Using lzlib %s\n", LZ_version() );
std::printf( "License GPLv2+: GNU GPL version 2 or later <http://gnu.org/licenses/gpl.html>\n" std::printf( "License GPLv2+: GNU GPL version 2 or later <http://gnu.org/licenses/gpl.html>\n"
"This is free software: you are free to change and redistribute it.\n" "This is free software: you are free to change and redistribute it.\n"
"There is NO WARRANTY, to the extent permitted by law.\n" ); "There is NO WARRANTY, to the extent permitted by law.\n" );
show_lzlib_version();
} }
@ -193,18 +208,7 @@ int check_lib()
std::printf( "warning: LZ_API_VERSION != LZ_api_version() (%u vs %u)\n", std::printf( "warning: LZ_API_VERSION != LZ_api_version() (%u vs %u)\n",
LZ_API_VERSION, LZ_api_version() ); } LZ_API_VERSION, LZ_api_version() ); }
#endif #endif
if( verbosity >= 1 ) if( verbosity >= 1 ) show_lzlib_version();
{
std::printf( "Using lzlib %s\n", LZ_version() );
#if !defined LZ_API_VERSION
std::fputs( "LZ_API_VERSION is not defined.\n", stdout );
#elif LZ_API_VERSION >= 1012
std::printf( "Using LZ_API_VERSION = %u\n", LZ_api_version() );
#else
std::printf( "Compiled with LZ_API_VERSION = %u. "
"Using an unknown LZ_API_VERSION\n", LZ_API_VERSION );
#endif
}
return retval; return retval;
} }
@ -222,7 +226,7 @@ const char * format_num3( long long num )
char * p = buf + bufsize - 1; // fill the buffer backwards char * p = buf + bufsize - 1; // fill the buffer backwards
*p = 0; // terminator *p = 0; // terminator
const bool negative = num < 0; const bool negative = num < 0;
if( num > 1024 || num < -1024 ) if( num > 9999 || num < -9999 )
{ {
char prefix = 0; // try binary first, then si char prefix = 0; // try binary first, then si
for( int i = 0; i < n && num != 0 && num % 1024 == 0; ++i ) for( int i = 0; i < n && num != 0 && num % 1024 == 0; ++i )
@ -269,7 +273,7 @@ long long getnum( const char * const arg, const char * const option_name,
if( !errno && tail[0] ) if( !errno && tail[0] )
{ {
const int factor = ( tail[1] == 'i' ) ? 1024 : 1000; const int factor = (tail[1] == 'i') ? 1024 : 1000;
int exponent = 0; // 0 = bad multiplier int exponent = 0; // 0 = bad multiplier
switch( tail[0] ) switch( tail[0] )
{ {
@ -354,8 +358,8 @@ long long parse_mtime( const char * arg, const char * const pn )
{ {
struct tm t; struct tm t;
t.tm_year = y - 1900; t.tm_mon = mo - 1; t.tm_mday = d; t.tm_year = y - 1900; t.tm_mon = mo - 1; t.tm_mday = d;
t.tm_hour = ( n >= 5 ) ? h : 0; t.tm_min = ( n >= 6 ) ? m : 0; t.tm_hour = (n >= 5) ? h : 0; t.tm_min = (n >= 6) ? m : 0;
t.tm_sec = ( n >= 7 ) ? s : 0; t.tm_isdst = -1; t.tm_sec = (n >= 7) ? s : 0; t.tm_isdst = -1;
errno = 0; errno = 0;
const long long mtime = std::mktime( &t ); const long long mtime = std::mktime( &t );
if( mtime != -1 || errno == 0 ) return mtime; // valid datetime if( mtime != -1 || errno == 0 ) return mtime; // valid datetime
@ -396,7 +400,7 @@ int hstat( const char * const filename, struct stat * const st,
bool nonempty_arg( const Arg_parser & parser, const int i ) bool nonempty_arg( const Arg_parser & parser, const int i )
{ return ( parser.code( i ) == 0 && !parser.argument( i ).empty() ); } { return parser.code( i ) == 0 && !parser.argument( i ).empty(); }
int open_instream( const std::string & name ) int open_instream( const std::string & name )
@ -582,7 +586,7 @@ int main( const int argc, const char * const argv[] )
{ opt_sol, "solid", Arg_parser::no }, { opt_sol, "solid", Arg_parser::no },
{ opt_un, "uncompressed", Arg_parser::no }, { opt_un, "uncompressed", Arg_parser::no },
{ opt_wn, "warn-newer", Arg_parser::no }, { opt_wn, "warn-newer", Arg_parser::no },
{ 0, 0, Arg_parser::no } }; { 0, 0, Arg_parser::no } };
const Arg_parser parser( argc, argv, options, true ); // in_order const Arg_parser parser( argc, argv, options, true ); // in_order
if( parser.error().size() ) // bad option if( parser.error().size() ) // bad option

35
tarlz.h
View file

@ -81,9 +81,9 @@ inline int decimal_digits( unsigned long long value )
inline bool dotdot_at_i( const char * const filename, const int i ) inline bool dotdot_at_i( const char * const filename, const int i )
{ {
return ( filename[i] == '.' && filename[i+1] == '.' && return filename[i] == '.' && filename[i+1] == '.' &&
( i == 0 || filename[i-1] == '/' ) && ( i == 0 || filename[i-1] == '/' ) &&
( filename[i+2] == 0 || filename[i+2] == '/' ) ); ( filename[i+2] == 0 || filename[i+2] == '/' );
} }
@ -244,10 +244,10 @@ class CRC32
uint32_t data[256]; // Table of CRCs of all 8-bit messages. uint32_t data[256]; // Table of CRCs of all 8-bit messages.
public: public:
CRC32( const bool castagnoli = false ) explicit CRC32( const bool castagnoli = false )
{ {
const unsigned cpol = 0x82F63B78U; // CRC32-C Castagnoli polynomial. const unsigned cpol = 0x82F63B78U; // CRC32-C Castagnoli polynomial
const unsigned ipol = 0xEDB88320U; // IEEE 802.3 Ethernet polynomial. const unsigned ipol = 0xEDB88320U; // IEEE 802.3 Ethernet polynomial
const unsigned poly = castagnoli ? cpol : ipol; const unsigned poly = castagnoli ? cpol : ipol;
for( unsigned n = 0; n < 256; ++n ) for( unsigned n = 0; n < 256; ++n )
@ -275,8 +275,7 @@ public:
uint32_t compute_crc( const uint8_t * const buffer, const int size ) const uint32_t compute_crc( const uint8_t * const buffer, const int size ) const
{ {
uint32_t crc = 0xFFFFFFFFU; uint32_t crc = 0xFFFFFFFFU;
for( int i = 0; i < size; ++i ) update_buf( crc, buffer, size );
crc = data[(crc^buffer[i])&0xFF] ^ ( crc >> 8 );
return crc ^ 0xFFFFFFFFU; return crc ^ 0xFFFFFFFFU;
} }
@ -322,8 +321,8 @@ enum {
inline bool isvalid_ds( const unsigned dictionary_size ) inline bool isvalid_ds( const unsigned dictionary_size )
{ return ( dictionary_size >= min_dictionary_size && { return dictionary_size >= min_dictionary_size &&
dictionary_size <= max_dictionary_size ); } dictionary_size <= max_dictionary_size; }
const uint8_t lzip_magic[4] = { 0x4C, 0x5A, 0x49, 0x50 }; // "LZIP" const uint8_t lzip_magic[4] = { 0x4C, 0x5A, 0x49, 0x50 }; // "LZIP"
@ -335,14 +334,13 @@ struct Lzip_header
// 4 version // 4 version
// 5 coded dictionary size // 5 coded dictionary size
bool check_magic() const bool check_magic() const { return std::memcmp( data, lzip_magic, 4 ) == 0; }
{ return ( std::memcmp( data, lzip_magic, 4 ) == 0 ); }
bool check_prefix( const int sz ) const // detect (truncated) header bool check_prefix( const int sz ) const // detect (truncated) header
{ {
for( int i = 0; i < sz && i < 4; ++i ) for( int i = 0; i < sz && i < 4; ++i )
if( data[i] != lzip_magic[i] ) return false; if( data[i] != lzip_magic[i] ) return false;
return ( sz > 0 ); return sz > 0;
} }
bool check_corrupt() const // detect corrupt header bool check_corrupt() const // detect corrupt header
@ -350,15 +348,15 @@ struct Lzip_header
int matches = 0; int matches = 0;
for( int i = 0; i < 4; ++i ) for( int i = 0; i < 4; ++i )
if( data[i] == lzip_magic[i] ) ++matches; if( data[i] == lzip_magic[i] ) ++matches;
return ( matches > 1 && matches < 4 ); return matches > 1 && matches < 4;
} }
uint8_t version() const { return data[4]; } uint8_t version() const { return data[4]; }
bool check_version() const { return ( data[4] == 1 ); } bool check_version() const { return data[4] == 1; }
unsigned dictionary_size() const unsigned dictionary_size() const
{ {
unsigned sz = ( 1 << ( data[5] & 0x1F ) ); unsigned sz = 1 << ( data[5] & 0x1F );
if( sz > min_dictionary_size ) if( sz > min_dictionary_size )
sz -= ( sz / 16 ) * ( ( data[5] >> 5 ) & 7 ); sz -= ( sz / 16 ) * ( ( data[5] >> 5 ) & 7 );
return sz; return sz;
@ -488,7 +486,7 @@ const char * const eclosa_msg = "Error closing archive";
const char * const eclosf_msg = "Error closing file"; const char * const eclosf_msg = "Error closing file";
const char * const nfound_msg = "Not found in archive."; const char * const nfound_msg = "Not found in archive.";
const char * const seek_msg = "Seek error"; const char * const seek_msg = "Seek error";
const char * const werr_msg = "Write error"; const char * const wr_err_msg = "Write error";
const char * const chdir_msg = "Error changing working directory"; const char * const chdir_msg = "Error changing working directory";
const char * const intdir_msg = "Failed to create intermediate directory"; const char * const intdir_msg = "Failed to create intermediate directory";
@ -521,7 +519,8 @@ void show_atpos_error( const char * const filename, const long long pos,
int compress( const Cl_options & cl_opts ); int compress( const Cl_options & cl_opts );
// defined in create.cc // defined in create.cc
bool copy_file( const int infd, const int outfd, const long long max_size = -1 ); bool copy_file( const int infd, const int outfd, const char * const filename,
const long long max_size = -1 );
bool writeblock_wrapper( const int outfd, const uint8_t * const buffer, bool writeblock_wrapper( const int outfd, const uint8_t * const buffer,
const int size ); const int size );
bool write_eoa_records( const int outfd, const bool compressed ); bool write_eoa_records( const int outfd, const bool compressed );

161
testsuite/check.sh
View file

@ -34,6 +34,7 @@ in_tar="${testdir}"/test.txt.tar
in_tar_lz="${testdir}"/test.txt.tar.lz in_tar_lz="${testdir}"/test.txt.tar.lz
inbad1="${testdir}"/test_bad1.txt inbad1="${testdir}"/test_bad1.txt
inbad2="${testdir}"/test_bad2.txt inbad2="${testdir}"/test_bad2.txt
em_lz="${testdir}"/em.lz
test3="${testdir}"/test3.tar test3="${testdir}"/test3.tar
test3_lz="${testdir}"/test3.tar.lz test3_lz="${testdir}"/test3.tar.lz
test3dir="${testdir}"/test3_dir.tar test3dir="${testdir}"/test3_dir.tar
@ -76,8 +77,8 @@ cyg_symlink() { [ ${lwarnc} = 0 ] &&
# tar_in_tlz1.tar.lz: 2 members (test.txt.tar test3.tar) 3 lzip members # tar_in_tlz1.tar.lz: 2 members (test.txt.tar test3.tar) 3 lzip members
# tar_in_tlz2.tar.lz: 2 members (test.txt.tar test3.tar) 5 lzip members # tar_in_tlz2.tar.lz: 2 members (test.txt.tar test3.tar) 5 lzip members
# ts_in_link.tar.lz: 4 symbolic links (link[1-4]) to / /dir/ dir/ dir(107/) # ts_in_link.tar.lz: 4 symbolic links (link[1-4]) to / /dir/ dir/ dir(107/)
# test_bad1.txt.tar.lz: truncated at offset 6000 (of 7495) # test_bad1.txt.tar.lz: truncated at offset 6000 (of 7459)
# test_bad2.txt.tar.lz: byte at offset 6000 changed from 0x56 to 0x46 # test_bad2.txt.tar.lz: byte at offset 6000 changed from 0x53 to 0x43
# test3.tar[.lz]: 3 members (foo bar baz) + 2 zeroed 512-byte blocks # test3.tar[.lz]: 3 members (foo bar baz) + 2 zeroed 512-byte blocks
# test3_dir.tar[.lz] like test3.tar but members /dir/foo /dir/bar /dir/baz # test3_dir.tar[.lz] like test3.tar but members /dir/foo /dir/bar /dir/baz
# test3_dot.tar.lz: 3 times 3 members ./foo ././bar ./././baz # test3_dot.tar.lz: 3 times 3 members ./foo ././bar ./././baz
@ -100,8 +101,6 @@ cyg_symlink() { [ ${lwarnc} = 0 ] &&
# test3_eoa3.tar.lz: test3.tar.lz with one zeroed block between foo and bar # test3_eoa3.tar.lz: test3.tar.lz with one zeroed block between foo and bar
# test3_eoa4.tar.lz: test3.tar.lz ended by extended header without EOA blocks # test3_eoa4.tar.lz: test3.tar.lz ended by extended header without EOA blocks
# test3_eoa5.tar.lz: test3.tar.lz split extended bar member, without EOA blocks # test3_eoa5.tar.lz: test3.tar.lz split extended bar member, without EOA blocks
# test3_em?.tar.lz: test3.tar.lz with one empty lzip member at each position
# test3_em6.tar.lz: test3.tar.lz preceded by four empty lzip members
# test3_gh?.tar: test3.tar with global header at each position # test3_gh?.tar: test3.tar with global header at each position
# test3_gh?.tar.lz: test3.tar.lz with global before bar split in 4 ways # test3_gh?.tar.lz: test3.tar.lz with global before bar split in 4 ways
# test3_gh5.tar.lz: test3.tar.lz with global in lzip member before foo # test3_gh5.tar.lz: test3.tar.lz with global in lzip member before foo
@ -235,9 +234,9 @@ done
rm -f out || framework_failure rm -f out || framework_failure
# test3 reference files for cmp # test3 reference files for cmp
cat "${testdir}"/rfoo > cfoo || framework_failure cp "${testdir}"/rfoo cfoo || framework_failure
cat "${testdir}"/rbar > cbar || framework_failure cp "${testdir}"/rbar cbar || framework_failure
cat "${testdir}"/rbaz > cbaz || framework_failure cp "${testdir}"/rbaz cbaz || framework_failure
# test --list and --extract test3 # test --list and --extract test3
rm -f foo bar baz || framework_failure rm -f foo bar baz || framework_failure
@ -492,9 +491,26 @@ for i in gh1 gh2 gh3 gh4 ; do
rm -f foo bar baz out || framework_failure rm -f foo bar baz out || framework_failure
done done
# test --list and --extract with empty lzip members, global headers and # test --list and --extract with empty lzip member
# extended tar members split among lzip members cat "${em_lz}" "${test3_lz}" > test3_em.tar.lz || framework_failure
for i in em1 em2 em3 em4 em5 em6 gh1 gh2 gh3 gh4 gh5 gh6 sm1 sm2 sm3 sm4 ; do "${TARLZ}" -q -tf test3_em.tar.lz > out
[ $? = 2 ] || test_failed $LINENO
"${TARLZ}" -tvf - < test3_em.tar.lz > out || test_failed $LINENO
diff -u vlist3 out || test_failed $LINENO
"${TARLZ}" -q -xf test3_em.tar.lz
[ $? = 2 ] || test_failed $LINENO
[ ! -e foo ] || test_failed $LINENO
[ ! -e bar ] || test_failed $LINENO
[ ! -e baz ] || test_failed $LINENO
"${TARLZ}" -xf - < test3_em.tar.lz || test_failed $LINENO
cmp cfoo foo || test_failed $LINENO
cmp cbar bar || test_failed $LINENO
cmp cbaz baz || test_failed $LINENO
rm -f foo bar baz || framework_failure
# test --list and --extract with global headers and extended tar members
# split among lzip members
for i in gh1 gh2 gh3 gh4 gh5 gh6 sm1 sm2 sm3 sm4 ; do
for j in 0 2 6 ; do for j in 0 2 6 ; do
"${TARLZ}" -n$j -tf "${testdir}"/test3_${i}.tar.lz > out || "${TARLZ}" -n$j -tf "${testdir}"/test3_${i}.tar.lz > out ||
test_failed $LINENO "$i $j" test_failed $LINENO "$i $j"
@ -518,14 +534,14 @@ rm -f list3 vlist3 || framework_failure
printf "\ntesting --concatenate..." printf "\ntesting --concatenate..."
# test --concatenate compressed # test --concatenate compressed
cat "${in}" > out.tar.lz || framework_failure # invalid tar.lz cp "${in}" out.tar.lz || framework_failure # invalid tar.lz
"${TARLZ}" -Aqf out.tar.lz "${test3_lz}" "${TARLZ}" -Aqf out.tar.lz "${test3_lz}"
[ $? = 2 ] || test_failed $LINENO [ $? = 2 ] || test_failed $LINENO
cat "${in_tar_lz}" > out.tar.lz || framework_failure cp "${in_tar_lz}" out.tar.lz || framework_failure
"${TARLZ}" -q --un -Af out.tar.lz "${test3_lz}" # contradictory ext "${TARLZ}" -q --un -Af out.tar.lz "${test3_lz}" # contradictory ext
[ $? = 1 ] || test_failed $LINENO [ $? = 1 ] || test_failed $LINENO
cmp "${in_tar_lz}" out.tar.lz || test_failed $LINENO cmp "${in_tar_lz}" out.tar.lz || test_failed $LINENO
cat "${in_tar_lz}" > out.tar.lz || framework_failure cp "${in_tar_lz}" out.tar.lz || framework_failure
"${TARLZ}" -Af out.tar.lz "${test3_lz}" || test_failed $LINENO "${TARLZ}" -Af out.tar.lz "${test3_lz}" || test_failed $LINENO
"${TARLZ}" -xf out.tar.lz || test_failed $LINENO "${TARLZ}" -xf out.tar.lz || test_failed $LINENO
cmp "${in}" test.txt || test_failed $LINENO cmp "${in}" test.txt || test_failed $LINENO
@ -547,12 +563,12 @@ cmp out.tar.lz aout.tar.lz || test_failed $LINENO
cmp "${in_tar_lz}" aout.tar.lz || test_failed $LINENO cmp "${in_tar_lz}" aout.tar.lz || test_failed $LINENO
"${TARLZ}" -A "${in_tar_lz}" "${test3_lz}" > aout.tar.lz || test_failed $LINENO "${TARLZ}" -A "${in_tar_lz}" "${test3_lz}" > aout.tar.lz || test_failed $LINENO
cmp out.tar.lz aout.tar.lz || test_failed $LINENO cmp out.tar.lz aout.tar.lz || test_failed $LINENO
cat "${eoa_lz}" > aout.tar.lz || framework_failure cp "${eoa_lz}" aout.tar.lz || framework_failure
"${TARLZ}" -Aqf aout.tar.lz "${in_tar}" # concatenate to empty archive "${TARLZ}" -Aqf aout.tar.lz "${in_tar}" # concatenate to empty archive
[ $? = 2 ] || test_failed $LINENO [ $? = 2 ] || test_failed $LINENO
"${TARLZ}" -Af aout.tar.lz "${in_tar_lz}" "${test3_lz}" || test_failed $LINENO "${TARLZ}" -Af aout.tar.lz "${in_tar_lz}" "${test3_lz}" || test_failed $LINENO
cmp out.tar.lz aout.tar.lz || test_failed $LINENO cmp out.tar.lz aout.tar.lz || test_failed $LINENO
cat "${in_tar_lz}" > aout.tar.lz || framework_failure cp "${in_tar_lz}" aout.tar.lz || framework_failure
"${TARLZ}" -Aqf aout.tar.lz "${test3_lz}" "${test3}" "${TARLZ}" -Aqf aout.tar.lz "${test3_lz}" "${test3}"
[ $? = 2 ] || test_failed $LINENO [ $? = 2 ] || test_failed $LINENO
cmp out.tar.lz aout.tar.lz || test_failed $LINENO cmp out.tar.lz aout.tar.lz || test_failed $LINENO
@ -566,14 +582,14 @@ cmp out.tar.lz aout.tar.lz || test_failed $LINENO
rm -f out.tar.lz aout.tar.lz || framework_failure rm -f out.tar.lz aout.tar.lz || framework_failure
# test --concatenate uncompressed # test --concatenate uncompressed
cat "${in}" > out.tar || framework_failure # invalid tar cp "${in}" out.tar || framework_failure # invalid tar
"${TARLZ}" -Aqf out.tar "${test3}" "${TARLZ}" -Aqf out.tar "${test3}"
[ $? = 2 ] || test_failed $LINENO [ $? = 2 ] || test_failed $LINENO
cat "${in_tar}" > out.tar || framework_failure cp "${in_tar}" out.tar || framework_failure
"${TARLZ}" -q -0 -Af out.tar "${test3}" # contradictory ext "${TARLZ}" -q -0 -Af out.tar "${test3}" # contradictory ext
[ $? = 1 ] || test_failed $LINENO [ $? = 1 ] || test_failed $LINENO
cmp "${in_tar}" out.tar || test_failed $LINENO cmp "${in_tar}" out.tar || test_failed $LINENO
cat "${in_tar}" > out.tar || framework_failure cp "${in_tar}" out.tar || framework_failure
"${TARLZ}" -Af out.tar "${test3}" || test_failed $LINENO "${TARLZ}" -Af out.tar "${test3}" || test_failed $LINENO
"${TARLZ}" -xf out.tar || test_failed $LINENO "${TARLZ}" -xf out.tar || test_failed $LINENO
cmp "${in}" test.txt || test_failed $LINENO cmp "${in}" test.txt || test_failed $LINENO
@ -595,12 +611,12 @@ cmp out.tar aout.tar || test_failed $LINENO
cmp "${in_tar}" aout.tar || test_failed $LINENO cmp "${in_tar}" aout.tar || test_failed $LINENO
"${TARLZ}" -A "${in_tar}" "${test3}" > aout.tar || test_failed $LINENO "${TARLZ}" -A "${in_tar}" "${test3}" > aout.tar || test_failed $LINENO
cmp out.tar aout.tar || test_failed $LINENO cmp out.tar aout.tar || test_failed $LINENO
cat "${eoa}" > aout.tar || framework_failure # concatenate to empty archive cp "${eoa}" aout.tar || framework_failure # concatenate to empty archive
"${TARLZ}" -Aqf aout.tar "${in_tar_lz}" "${TARLZ}" -Aqf aout.tar "${in_tar_lz}"
[ $? = 2 ] || test_failed $LINENO [ $? = 2 ] || test_failed $LINENO
"${TARLZ}" -Af aout.tar "${in_tar}" "${test3}" || test_failed $LINENO "${TARLZ}" -Af aout.tar "${in_tar}" "${test3}" || test_failed $LINENO
cmp out.tar aout.tar || test_failed $LINENO cmp out.tar aout.tar || test_failed $LINENO
cat "${in_tar}" > aout.tar || framework_failure cp "${in_tar}" aout.tar || framework_failure
"${TARLZ}" -Aqf aout.tar "${test3}" "${test3_lz}" "${TARLZ}" -Aqf aout.tar "${test3}" "${test3_lz}"
[ $? = 2 ] || test_failed $LINENO [ $? = 2 ] || test_failed $LINENO
cmp out.tar aout.tar || test_failed $LINENO cmp out.tar aout.tar || test_failed $LINENO
@ -616,13 +632,13 @@ rm -f out.tar aout.tar || framework_failure
printf "\ntesting --create..." printf "\ntesting --create..."
# test --create # test --create
cat "${in}" > test.txt || framework_failure cp "${in}" test.txt || framework_failure
"${TARLZ}" --warn-newer -0 -cf out.tar.lz test.txt || test_failed $LINENO "${TARLZ}" --warn-newer -0 -cf out.tar.lz test.txt || test_failed $LINENO
is_compressed out.tar.lz || test_failed $LINENO is_compressed out.tar.lz || test_failed $LINENO
rm -f test.txt || framework_failure rm -f test.txt || framework_failure
"${TARLZ}" -xf out.tar.lz --missing-crc || test_failed $LINENO "${TARLZ}" -xf out.tar.lz --missing-crc || test_failed $LINENO
cmp "${in}" test.txt || test_failed $LINENO cmp "${in}" test.txt || test_failed $LINENO
cat "${in}" > test.txt || framework_failure cp "${in}" test.txt || framework_failure
"${TARLZ}" --warn-newer --un -cf out.tar test.txt || test_failed $LINENO "${TARLZ}" --warn-newer --un -cf out.tar test.txt || test_failed $LINENO
is_uncompressed out.tar || test_failed $LINENO is_uncompressed out.tar || test_failed $LINENO
rm -f test.txt || framework_failure rm -f test.txt || framework_failure
@ -630,9 +646,9 @@ rm -f test.txt || framework_failure
cmp "${in}" test.txt || test_failed $LINENO cmp "${in}" test.txt || test_failed $LINENO
rm -f test.txt out.tar out.tar.lz || framework_failure rm -f test.txt out.tar out.tar.lz || framework_failure
cat cfoo > foo || framework_failure cp cfoo foo || framework_failure
rm -f bar || framework_failure rm -f bar || framework_failure
cat cbaz > baz || framework_failure cp cbaz baz || framework_failure
"${TARLZ}" -0 -q -cf out.tar.lz foo bar baz "${TARLZ}" -0 -q -cf out.tar.lz foo bar baz
[ $? = 1 ] || test_failed $LINENO [ $? = 1 ] || test_failed $LINENO
rm -f foo bar baz || framework_failure rm -f foo bar baz || framework_failure
@ -648,9 +664,9 @@ rm -f foo bar baz || framework_failure
[ ! -e baz ] || test_failed $LINENO [ ! -e baz ] || test_failed $LINENO
rm -f out.tar.lz || framework_failure rm -f out.tar.lz || framework_failure
cat cfoo > foo || framework_failure cp cfoo foo || framework_failure
cat cbar > bar || framework_failure cp cbar bar || framework_failure
cat cbaz > baz || framework_failure cp cbaz baz || framework_failure
"${TARLZ}" -0 -cf out.tar.lz foo bar baz --out-slots=1 || test_failed $LINENO "${TARLZ}" -0 -cf out.tar.lz foo bar baz --out-slots=1 || test_failed $LINENO
"${TARLZ}" -0 -q -cf aout.tar.lz foo bar aout.tar.lz baz || test_failed $LINENO "${TARLZ}" -0 -q -cf aout.tar.lz foo bar aout.tar.lz baz || test_failed $LINENO
cmp out.tar.lz aout.tar.lz || test_failed $LINENO # test reproducible cmp out.tar.lz aout.tar.lz || test_failed $LINENO # test reproducible
@ -709,9 +725,9 @@ rm -rf dir1 || framework_failure
rm -f out.tar.lz aout.tar.lz || framework_failure rm -f out.tar.lz aout.tar.lz || framework_failure
# test --create --exclude # test --create --exclude
cat cfoo > foo || framework_failure cp cfoo foo || framework_failure
cat cbar > bar || framework_failure cp cbar bar || framework_failure
cat cbaz > baz || framework_failure cp cbaz baz || framework_failure
"${TARLZ}" -0 -cf out.tar.lz foo bar baz --exclude 'ba?' || test_failed $LINENO "${TARLZ}" -0 -cf out.tar.lz foo bar baz --exclude 'ba?' || test_failed $LINENO
rm -f foo bar baz || framework_failure rm -f foo bar baz || framework_failure
"${TARLZ}" -xf out.tar.lz || test_failed $LINENO "${TARLZ}" -xf out.tar.lz || test_failed $LINENO
@ -719,9 +735,9 @@ cmp cfoo foo || test_failed $LINENO
[ ! -e bar ] || test_failed $LINENO [ ! -e bar ] || test_failed $LINENO
[ ! -e baz ] || test_failed $LINENO [ ! -e baz ] || test_failed $LINENO
rm -f out.tar.lz foo bar baz || framework_failure rm -f out.tar.lz foo bar baz || framework_failure
cat cfoo > foo || framework_failure cp cfoo foo || framework_failure
cat cbar > bar || framework_failure cp cbar bar || framework_failure
cat cbaz > baz || framework_failure cp cbaz baz || framework_failure
"${TARLZ}" -cf out.tar foo bar baz --exclude 'ba*' || test_failed $LINENO "${TARLZ}" -cf out.tar foo bar baz --exclude 'ba*' || test_failed $LINENO
is_uncompressed out.tar || test_failed $LINENO is_uncompressed out.tar || test_failed $LINENO
rm -f foo bar baz || framework_failure rm -f foo bar baz || framework_failure
@ -746,9 +762,14 @@ dates='@-9223372036854775808 @-9223372036854775807
2242-03-16T12:56:31 2242-03-16T12:56:32 @8589934591 @8589934592 2242-03-16T12:56:31 2242-03-16T12:56:32 @8589934591 @8589934592
9999-12-31T23:59:58 9999-12-31T23:59:59 9999-12-31T23:59:58 9999-12-31T23:59:59
2147483647-12-31T23:59:59 @9223372036854775807' 2147483647-12-31T23:59:59 @9223372036854775807'
touch -d 2022-01-05T12:22:13 bar || framework_failure touch -d 2022-01-05T12:22:13 bar >/dev/null 2>&1 ||
touch -d '2022-01-05 12:22:13' bar >/dev/null 2>&1 ||
touch bar || framework_failure
for i in ${dates} @-8Ei '2017-10-01 09:00:00' '2017-10-1 9:0:0' \ for i in ${dates} @-8Ei '2017-10-01 09:00:00' '2017-10-1 9:0:0' \
'2017-10-01 09:00' '2017-10-01 09' 2017-10-01 ./bar ; do '2017-10-01 09:00' '2017-10-01 09' 2017-10-01 ./bar ; do
# Skip a time stamp $i if it's out of range for this platform,
# or if it uses a notation that this platform does not recognize.
[ "$i" = "./bar" ] || touch -d "$i" foo >/dev/null 2>&1 || continue
touch foo || framework_failure touch foo || framework_failure
"${TARLZ}" -cf out.tar --mtime="$i" foo || test_failed $LINENO "$i" "${TARLZ}" -cf out.tar --mtime="$i" foo || test_failed $LINENO "$i"
is_uncompressed out.tar || test_failed $LINENO "$i" is_uncompressed out.tar || test_failed $LINENO "$i"
@ -761,11 +782,7 @@ done
rm -f out.tar foo bar || framework_failure rm -f out.tar foo bar || framework_failure
mkdir dir || framework_failure mkdir dir || framework_failure
for i in ${dates} ; do for i in ${dates} ; do touch -d "$i" "dir/f$i" >/dev/null 2>&1 ; done
# Skip a time stamp $i if it's out of range for this platform,
# of if it uses a notation that this platform does not recognize.
touch -d "$i" "dir/f$i" >/dev/null 2>&1 || continue
done
"${TARLZ}" -cf out.tar dir || test_failed $LINENO "${TARLZ}" -cf out.tar dir || test_failed $LINENO
is_uncompressed out.tar || test_failed $LINENO is_uncompressed out.tar || test_failed $LINENO
"${TARLZ}" -df out.tar || test_failed $LINENO "${TARLZ}" -df out.tar || test_failed $LINENO
@ -814,14 +831,14 @@ rm -f out.tar aout.tar foo bar baz || framework_failure
printf "\ntesting --delete..." printf "\ntesting --delete..."
# test --delete # test --delete
cat "${in}" > out.tar || framework_failure # invalid tar cp "${in}" out.tar || framework_failure # invalid tar
"${TARLZ}" -q -f out.tar --delete foo "${TARLZ}" -q -f out.tar --delete foo
[ $? = 2 ] || test_failed $LINENO [ $? = 2 ] || test_failed $LINENO
rm -f out.tar || framework_failure rm -f out.tar || framework_failure
cat "${in}" > out.tar.lz || framework_failure # invalid tar.lz cp "${in}" out.tar.lz || framework_failure # invalid tar.lz
"${TARLZ}" -q -f out.tar.lz --delete foo "${TARLZ}" -q -f out.tar.lz --delete foo
[ $? = 2 ] || test_failed $LINENO [ $? = 2 ] || test_failed $LINENO
cat "${in_lz}" > out.tar.lz || framework_failure # invalid tar.lz cp "${in_lz}" out.tar.lz || framework_failure # invalid tar.lz
"${TARLZ}" -q -f out.tar.lz --delete foo "${TARLZ}" -q -f out.tar.lz --delete foo
[ $? = 2 ] || test_failed $LINENO [ $? = 2 ] || test_failed $LINENO
rm -f out.tar.lz || framework_failure rm -f out.tar.lz || framework_failure
@ -881,10 +898,10 @@ for e in "" .lz ; do
done done
# test --delete individual member after collective member # test --delete individual member after collective member
cat cfoo > foo || framework_failure cp cfoo foo || framework_failure
cat cbar > bar || framework_failure cp cbar bar || framework_failure
cat cbaz > baz || framework_failure cp cbaz baz || framework_failure
cat "${in}" > test.txt || framework_failure cp "${in}" test.txt || framework_failure
"${TARLZ}" -0 -cf out.tar.lz foo bar baz --asolid || test_failed $LINENO "${TARLZ}" -0 -cf out.tar.lz foo bar baz --asolid || test_failed $LINENO
"${TARLZ}" -0 -rf out.tar.lz test.txt || test_failed $LINENO "${TARLZ}" -0 -rf out.tar.lz test.txt || test_failed $LINENO
rm -f foo bar baz test.txt || framework_failure rm -f foo bar baz test.txt || framework_failure
@ -901,20 +918,16 @@ cmp cbaz baz || test_failed $LINENO
rm -f out.tar.lz foo bar baz test.txt || framework_failure rm -f out.tar.lz foo bar baz test.txt || framework_failure
# test --delete with empty lzip member, global header # test --delete with empty lzip member, global header
for i in 1 2 3 4 5 6 ; do "${TARLZ}" -q -f test3_em.tar.lz --delete foo
cat "${testdir}"/test3_em${i}.tar.lz > out.tar.lz || framework_failure [ $? = 2 ] || test_failed $LINENO
for j in foo bar baz ; do rm -f test3_em.tar.lz || framework_failure
"${TARLZ}" -f out.tar.lz --delete $j || test_failed $LINENO "$i $j" cp "${testdir}"/test3_gh5.tar.lz out.tar.lz || framework_failure
done
rm -f out.tar.lz || framework_failure
done
cat "${testdir}"/test3_gh5.tar.lz > out.tar.lz || framework_failure
for i in foo bar baz ; do for i in foo bar baz ; do
"${TARLZ}" -f out.tar.lz --delete $i || test_failed $LINENO $i "${TARLZ}" -f out.tar.lz --delete $i || test_failed $LINENO $i
done done
rm -f out.tar.lz || framework_failure rm -f out.tar.lz || framework_failure
for i in 1 2 3 4 ; do for i in 1 2 3 4 ; do
cat "${testdir}"/test3_gh${i}.tar > out.tar || framework_failure cp "${testdir}"/test3_gh${i}.tar out.tar || framework_failure
for j in foo bar baz ; do for j in foo bar baz ; do
"${TARLZ}" -f out.tar --delete $j || test_failed $LINENO "$i $j" "${TARLZ}" -f out.tar --delete $j || test_failed $LINENO "$i $j"
done done
@ -935,9 +948,9 @@ rm -f dummy_slink dummy_link dummy_file || framework_failure
# #
if [ "${ln_works}" = yes ] ; then if [ "${ln_works}" = yes ] ; then
mkdir dir || framework_failure mkdir dir || framework_failure
cat cfoo > dir/foo || framework_failure cp cfoo dir/foo || framework_failure
cat cbar > dir/bar || framework_failure cp cbar dir/bar || framework_failure
cat cbaz > dir/baz || framework_failure cp cbaz dir/baz || framework_failure
ln -s dir dir_link || framework_failure ln -s dir dir_link || framework_failure
"${TARLZ}" -0 -c dir_link > out1 || test_failed $LINENO "${TARLZ}" -0 -c dir_link > out1 || test_failed $LINENO
is_compressed out1 || test_failed $LINENO is_compressed out1 || test_failed $LINENO
@ -968,9 +981,9 @@ fi
printf "\ntesting --append..." printf "\ntesting --append..."
# test --append compressed # test --append compressed
cat cfoo > foo || framework_failure cp cfoo foo || framework_failure
cat cbar > bar || framework_failure cp cbar bar || framework_failure
cat cbaz > baz || framework_failure cp cbaz baz || framework_failure
"${TARLZ}" -0 -cf out.tar.lz foo bar baz --out-slots=1024 || test_failed $LINENO "${TARLZ}" -0 -cf out.tar.lz foo bar baz --out-slots=1024 || test_failed $LINENO
"${TARLZ}" -0 -cf nout.tar.lz foo bar baz --no-solid || test_failed $LINENO "${TARLZ}" -0 -cf nout.tar.lz foo bar baz --no-solid || test_failed $LINENO
"${TARLZ}" -0 -cf aout.tar.lz foo || test_failed $LINENO "${TARLZ}" -0 -cf aout.tar.lz foo || test_failed $LINENO
@ -994,7 +1007,7 @@ cmp out.tar.lz aout.tar.lz || test_failed $LINENO
"${TARLZ}" --un -q -rf aout.tar.lz foo bar baz # contradictory ext "${TARLZ}" --un -q -rf aout.tar.lz foo bar baz # contradictory ext
[ $? = 1 ] || test_failed $LINENO [ $? = 1 ] || test_failed $LINENO
cmp out.tar.lz aout.tar.lz || test_failed $LINENO cmp out.tar.lz aout.tar.lz || test_failed $LINENO
cat "${eoa_lz}" > aout.tar.lz || framework_failure # append to empty archive cp "${eoa_lz}" aout.tar.lz || framework_failure # append to empty archive
"${TARLZ}" -0 -rf aout.tar.lz foo bar baz || test_failed $LINENO "${TARLZ}" -0 -rf aout.tar.lz foo bar baz || test_failed $LINENO
cmp out.tar.lz aout.tar.lz || test_failed $LINENO cmp out.tar.lz aout.tar.lz || test_failed $LINENO
rm -f out.tar.lz aout.tar.lz || framework_failure rm -f out.tar.lz aout.tar.lz || framework_failure
@ -1028,7 +1041,7 @@ cmp out.tar aout.tar || test_failed $LINENO
"${TARLZ}" -0 -q -rf aout.tar foo bar baz # contradictory ext "${TARLZ}" -0 -q -rf aout.tar foo bar baz # contradictory ext
[ $? = 1 ] || test_failed $LINENO [ $? = 1 ] || test_failed $LINENO
cmp out.tar aout.tar || test_failed $LINENO cmp out.tar aout.tar || test_failed $LINENO
cat "${eoa}" > aout.tar || framework_failure # append to empty archive cp "${eoa}" aout.tar || framework_failure # append to empty archive
"${TARLZ}" -rf aout.tar foo bar baz || test_failed $LINENO "${TARLZ}" -rf aout.tar foo bar baz || test_failed $LINENO
cmp out.tar aout.tar || test_failed $LINENO cmp out.tar aout.tar || test_failed $LINENO
rm -f out.tar aout.tar || framework_failure rm -f out.tar aout.tar || framework_failure
@ -1036,7 +1049,7 @@ rm -f out.tar aout.tar || framework_failure
# test --append to solid archive # test --append to solid archive
"${TARLZ}" --solid -q -0 -cf out.tar.lz "${in}" foo bar || test_failed $LINENO "${TARLZ}" --solid -q -0 -cf out.tar.lz "${in}" foo bar || test_failed $LINENO
"${TARLZ}" -q -tf out.tar.lz || test_failed $LINENO # compressed seekable "${TARLZ}" -q -tf out.tar.lz || test_failed $LINENO # compressed seekable
cat out.tar.lz > aout.tar.lz || framework_failure cp out.tar.lz aout.tar.lz || framework_failure
for i in --asolid --bsolid --dsolid --solid -0 ; do for i in --asolid --bsolid --dsolid --solid -0 ; do
"${TARLZ}" $i -q -rf out.tar.lz baz "${TARLZ}" $i -q -rf out.tar.lz baz
[ $? = 2 ] || test_failed $LINENO $i [ $? = 2 ] || test_failed $LINENO $i
@ -1083,7 +1096,7 @@ if [ "${ln_works}" = yes ] ; then
mkdir dir1 || framework_failure mkdir dir1 || framework_failure
mkdir dir1/dir2 || framework_failure mkdir dir1/dir2 || framework_failure
mkdir dir1/dir2/dir3 || framework_failure mkdir dir1/dir2/dir3 || framework_failure
cat "${in}" > dir1/dir2/dir3/in || framework_failure cp "${in}" dir1/dir2/dir3/in || framework_failure
ln dir1/dir2/dir3/in dir1/dir2/dir3/"${name_100}" || framework_failure ln dir1/dir2/dir3/in dir1/dir2/dir3/"${name_100}" || framework_failure
ln dir1/dir2/dir3/in "${path_100}" || framework_failure ln dir1/dir2/dir3/in "${path_100}" || framework_failure
ln dir1/dir2/dir3/in "${path_106}" || framework_failure ln dir1/dir2/dir3/in "${path_106}" || framework_failure
@ -1160,20 +1173,20 @@ rm -f foo || framework_failure
printf "\ntesting --compress..." printf "\ntesting --compress..."
cat cfoo > foo || framework_failure cp cfoo foo || framework_failure
cat cbar > bar || framework_failure cp cbar bar || framework_failure
cat cbaz > baz || framework_failure cp cbaz baz || framework_failure
cat "${in}" > test.txt || framework_failure cp "${in}" test.txt || framework_failure
"${TARLZ}" -cf out.tar test.txt foo bar baz test.txt || test_failed $LINENO "${TARLZ}" -cf out.tar test.txt foo bar baz test.txt || test_failed $LINENO
"${TARLZ}" -cf out3.tar foo bar baz || test_failed $LINENO "${TARLZ}" -cf out3.tar foo bar baz || test_failed $LINENO
cat out.tar > outz.tar || framework_failure cp out.tar outz.tar || framework_failure
cat out3.tar > out3z.tar || framework_failure cp out3.tar out3z.tar || framework_failure
# #
"${TARLZ}" -0 -z outz.tar out3z.tar || test_failed $LINENO "${TARLZ}" -0 -z outz.tar out3z.tar || test_failed $LINENO
"${TARLZ}" -q -tf outz.tar.lz || test_failed $LINENO "${TARLZ}" -q -tf outz.tar.lz || test_failed $LINENO
"${TARLZ}" -q -tf out3z.tar.lz || test_failed $LINENO "${TARLZ}" -q -tf out3z.tar.lz || test_failed $LINENO
cat outz.tar.lz > out || test_failed $LINENO cp outz.tar.lz out || test_failed $LINENO
cat out3z.tar.lz > out3 || test_failed $LINENO cp out3z.tar.lz out3 || test_failed $LINENO
rm -f out3z.tar.lz || framework_failure rm -f out3z.tar.lz || framework_failure
"${TARLZ}" -q -0 -z outz.tar out3z.tar # outz.tar.lz exists "${TARLZ}" -q -0 -z outz.tar out3z.tar # outz.tar.lz exists
[ $? = 1 ] || test_failed $LINENO [ $? = 1 ] || test_failed $LINENO
@ -1271,7 +1284,7 @@ rm -f truncated.tar || framework_failure
# test --delete with split 'bar' tar member # test --delete with split 'bar' tar member
for i in 1 2 3 4 ; do for i in 1 2 3 4 ; do
cat "${testdir}"/test3_sm${i}.tar.lz > out.tar.lz || framework_failure cp "${testdir}"/test3_sm${i}.tar.lz out.tar.lz || framework_failure
for j in bar baz ; do for j in bar baz ; do
"${TARLZ}" -q -f out.tar.lz --delete $j "${TARLZ}" -q -f out.tar.lz --delete $j
[ $? = 2 ] || test_failed $LINENO "$i $j" [ $? = 2 ] || test_failed $LINENO "$i $j"

BIN
testsuite/em.lz Normal file
View file

Binary file not shown.

BIN
testsuite/tar_in_tlz1.tar.lz
View file

Binary file not shown.

BIN
testsuite/tar_in_tlz2.tar.lz
View file

Binary file not shown.

6
testsuite/test.txt
View file

@ -1,8 +1,7 @@
GNU GENERAL PUBLIC LICENSE GNU GENERAL PUBLIC LICENSE
Version 2, June 1991 Version 2, June 1991
Copyright (C) 1989, 1991 Free Software Foundation, Inc., Copyright (C) 1989, 1991 Free Software Foundation, Inc. <http://fsf.org/>
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Everyone is permitted to copy and distribute verbatim copies Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed. of this license document, but changing it is not allowed.
@ -339,8 +338,7 @@ Public License instead of this License.
GNU GENERAL PUBLIC LICENSE GNU GENERAL PUBLIC LICENSE
Version 2, June 1991 Version 2, June 1991
Copyright (C) 1989, 1991 Free Software Foundation, Inc., Copyright (C) 1989, 1991 Free Software Foundation, Inc. <http://fsf.org/>
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Everyone is permitted to copy and distribute verbatim copies Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed. of this license document, but changing it is not allowed.

BIN
testsuite/test.txt.lz
View file

Binary file not shown.

BIN
testsuite/test.txt.tar
View file

Binary file not shown.

BIN
testsuite/test.txt.tar.lz
View file

Binary file not shown.

BIN
testsuite/test3_em1.tar.lz
View file

Binary file not shown.

BIN
testsuite/test3_em2.tar.lz
View file

Binary file not shown.

BIN
testsuite/test3_em3.tar.lz
View file

Binary file not shown.

BIN
testsuite/test3_em4.tar.lz
View file

Binary file not shown.

BIN
testsuite/test3_em5.tar.lz
View file

Binary file not shown.

BIN
testsuite/test3_em6.tar.lz
View file

Binary file not shown.

7
testsuite/test_bad1.txt
View file

@ -1,8 +1,7 @@
GNU GENERAL PUBLIC LICENSE GNU GENERAL PUBLIC LICENSE
Version 2, June 1991 Version 2, June 1991
Copyright (C) 1989, 1991 Free Software Foundation, Inc., Copyright (C) 1989, 1991 Free Software Foundation, Inc. <http://fsf.org/>
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Everyone is permitted to copy and distribute verbatim copies Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed. of this license document, but changing it is not allowed.
@ -304,4 +303,6 @@ the "copyright" line and a pointer to where the full notice is found.
GNU General Public License for more details. GNU General Public License for more details.
You should have received a copy of the GNU General Public License You should have received a copy of the GNU General Public License
along with this program along with this program. If not, see <http://www.gnu.org/licenses/>.
Also add information on how to contact you by electronic and pa

BIN
testsuite/test_bad1.txt.tar
View file

Binary file not shown.

BIN
testsuite/test_bad1.txt.tar.lz
View file

Binary file not shown.

19
testsuite/test_bad2.txt
View file

@ -1,8 +1,7 @@
GNU GENERAL PUBLIC LICENSE GNU GENERAL PUBLIC LICENSE
Version 2, June 1991 Version 2, June 1991
Copyright (C) 1989, 1991 Free Software Foundation, Inc., Copyright (C) 1989, 1991 Free Software Foundation, Inc. <http://fsf.org/>
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Everyone is permitted to copy and distribute verbatim copies Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed. of this license document, but changing it is not allowed.
@ -304,17 +303,7 @@ the "copyright" line and a pointer to where the full notice is found.
GNU General Public License for more details. GNU General Public License for more details.
You should have received a copy of the GNU General Public License You should have received a copy of the GNU General Public License
along with this program. Ifodifnperived from the Progr"work based on therogrdifneneraeuse of software generally. along with this program. If not, see <http://www.gnu.org/licenses/>.
NO WARRANTY Also add information on how to contact you by electronic and papeLicost ordinram iThis Ysequence ofttp
erwise to cod the ITING THE any e licenntradict thel cense, CH DAM
11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT X FR TO NO WARRANTY
11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT X FR TO NO WARRANTY
11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARQIS NO WARRATHERE IS NO WARRANTY
FOR THE
This ee

BIN
testsuite/test_bad2.txt.tar.lz
View file

Binary file not shown.