diff options
| -rw-r--r-- | ChangeLog | 51 | ||||
| -rw-r--r-- | Makefile.in | 7 | ||||
| -rw-r--r-- | NEWS | 19 | ||||
| -rw-r--r-- | alone_to_lz.cc | 6 | ||||
| -rw-r--r-- | byte_repair.cc | 91 | ||||
| -rwxr-xr-x | configure | 2 | ||||
| -rw-r--r-- | decoder.cc | 28 | ||||
| -rw-r--r-- | decoder.h | 20 | ||||
| -rw-r--r-- | doc/lziprecover.1 | 8 | ||||
| -rw-r--r-- | doc/lziprecover.info | 381 | ||||
| -rw-r--r-- | doc/lziprecover.texi | 259 | ||||
| -rw-r--r-- | dump_remove.cc | 28 | ||||
| -rw-r--r-- | fec.h | 1 | ||||
| -rw-r--r-- | fec_create.cc | 16 | ||||
| -rw-r--r-- | fec_repair.cc | 21 | ||||
| -rw-r--r-- | gf16.cc | 2 | ||||
| -rw-r--r-- | gf8.cc | 4 | ||||
| -rw-r--r-- | list.cc | 14 | ||||
| -rw-r--r-- | lunzcrash.cc | 42 | ||||
| -rw-r--r-- | lzip.h | 44 | ||||
| -rw-r--r-- | lzip_index.cc | 16 | ||||
| -rw-r--r-- | lzip_index.h | 8 | ||||
| -rw-r--r-- | main.cc | 87 | ||||
| -rw-r--r-- | merge.cc | 86 | ||||
| -rw-r--r-- | mtester.cc | 48 | ||||
| -rw-r--r-- | mtester.h | 19 | ||||
| -rw-r--r-- | nrep_stats.cc | 4 | ||||
| -rw-r--r-- | range_dec.cc | 15 | ||||
| -rw-r--r-- | recursive.cc | 2 | ||||
| -rw-r--r-- | reproduce.cc | 29 | ||||
| -rw-r--r-- | split.cc | 14 | ||||
| -rwxr-xr-x | testsuite/check.sh | 339 | ||||
| -rw-r--r-- | testsuite/fox6.lz | bin | 480 -> 0 bytes | |||
| -rw-r--r-- | testsuite/fox6_b1nz.lz | bin | 0 -> 480 bytes | |||
| -rw-r--r-- | testsuite/fox6_nz.lz | bin | 480 -> 0 bytes | |||
| -rw-r--r-- | testsuite/fox_nz.lz | bin | 0 -> 80 bytes | |||
| -rw-r--r-- | testsuite/test_3m.txt.lz.md5 | 1 | ||||
| -rw-r--r-- | testsuite/test_em.txt.lz | bin | 13950 -> 0 bytes | |||
| -rw-r--r-- | unzcrash.cc | 8 |
39 files changed, 978 insertions, 742 deletions
@@ -1,3 +1,15 @@ +2024-11-18 Antonio Diaz Diaz <antonio@gnu.org> + + * Version 1.25-rc1 released. + * byte_repair.cc: Repair a nonzero first LZMA byte. + * Integrate options '--ignore-empty' and '--ignore-nonzero' into + '-i, --ignore-errors'. + * merge.cc (copy_file): Add name arguments, use 'show_file_error'. + * lziprecover.texi: New chapter 'Syntax of command-line arguments'. + * check.sh: Use 'cp' instead of 'cat'. + * testsuite: Add fox_nz.lz, fox6_b1nz.lz. + Remove fox6.lz, fox6_nz.lz, test_em.txt.lz, test_3m.txt.lz.md5. + 2024-10-01 Antonio Diaz Diaz <antonio@gnu.org> * Version 1.25-pre1 released. @@ -7,7 +19,7 @@ * New options '--ignore-empty' and '--ignore-nonzero'. * Rename option '--clear-marking' to '--nonzero-repair'. * Remove options '--empty-error' and '--marking-error'. - * Remove decompression support for Sync Flush marker. + * decoder.cc (decode_member): Remove support for Sync Flush marker. * testsuite: Require lzip/clzip. Add fox6_nz.lz. Remove fox6_mark.lz. 2024-01-20 Antonio Diaz Diaz <antonio@gnu.org> @@ -134,11 +146,10 @@ * repair.cc: Repair a damaged dictionary size in the header. * repair.cc: Try bytes at offsets 7 to 11 first. * Decompression time has been reduced by 2%. - * main.cc (decompress): Print up to 6 bytes of trailing data when - '-tvvvv' is specified. - * decoder.cc (verify_trailer): Remove test of final code. * main.cc (main): Delete '--output' file if infd is a terminal. - * main.cc (main): Don't use stdin more than once. + (main): Don't use stdin more than once. + (decompress): Print 6 bytes of trailing data at verbosity level 4. + * decoder.cc (verify_trailer): Remove test of final code. * Use 'close_and_set_permissions' and 'set_signals' in all modes. * range_dec.cc (list_file): Show dictionary size and size of trailing data (if any) with '-lv'. @@ -154,8 +165,7 @@ * lziprecover.texi: New chapter 'Trailing data'. * configure: Avoid warning on some shells when testing for g++. * Makefile.in: Detect the existence of install-info. - * check.sh: Don't check error messages. - * check.sh: A POSIX shell is required to run the tests. + * check.sh: Require a POSIX shell. Don't check error messages. 2015-05-28 Antonio Diaz Diaz <antonio@gnu.org> @@ -192,11 +202,11 @@ * Option '-l, --list' now accepts more than one file. * Decompression time has been reduced by 12%. * split.cc: Use as few digits as possible in file names. - * split.cc: In verbose mode show names of files being created. + In verbose mode show names of files being created. * main.cc (show_header): Show header version if verbosity >= 4. + (main): Use 'setmode' instead of '_setmode' on Windows and OS/2. * configure: Options now accept a separate argument. * Makefile.in: New targets 'install-as-lzip' and 'install-bin'. - * main.cc: Use 'setmode' instead of '_setmode' on Windows and OS/2. 2012-02-24 Antonio Diaz Diaz <ant_diaz@teleline.es> @@ -207,8 +217,7 @@ * lziprecover.cc: Rename to main.cc. * New files merge.cc, repair.cc, split.cc, and range_dec.cc. * main.cc: Add decompressor options (-c, -d, -k, -t) so that an - external decompressor is not needed for recovery nor for - "make check". + external decompressor is not needed for recovery and 'make check'. * New option '-D, --range-decompress', which extracts a range of bytes decompressing only the members containing the desired data. * New option '-l, --list', which prints correct total file sizes @@ -223,25 +232,23 @@ * Version 1.12 released. * lziprecover.cc: If '-v' is not specified show errors only. * unzcrash.cc: Use Arg_parser. - * unzcrash.cc: New options '-b, --bits', '-p, --position', and - '-s, --size'. + New options '-b, --bits', '-p, --position', and '-s, --size'. 2010-09-16 Antonio Diaz Diaz <ant_diaz@teleline.es> * Version 1.11 released. - * lziprecover.cc: New option '-m, --merge', which tries to produce a - correct file by merging the good parts of two or more damaged copies. - * lziprecover.cc: New option '-R, --repair' for repairing a - 1-byte error in single-member files. * decoder.cc (decode_member): Detect file errors earlier to improve efficiency of lziprecover's new repair capability. This change also prevents (harmless) access to uninitialized memory when decompressing a corrupt file. - * lziprecover.cc: New options '-f, --force' and '-o, --output'. - * lziprecover.cc: New option '-s, --split' to select the until now - only operation of splitting multimember files. - * lziprecover.cc: If no operation is specified, warn the user and do - nothing. + * lziprecover.cc: New option '-m, --merge', which tries to produce a + correct file by merging the good parts of two or more damaged copies. + New option '-R, --repair' for repairing a 1-byte error in + single-member files. + New options '-f, --force' and '-o, --output'. + New option '-s, --split' to select the until now only operation of + splitting multimember files. + If no operation is specified, warn the user and do nothing. 2009-06-22 Antonio Diaz Diaz <ant_diaz@teleline.es> diff --git a/Makefile.in b/Makefile.in index fead3d2..11278fa 100644 --- a/Makefile.in +++ b/Makefile.in @@ -2,8 +2,8 @@ DISTNAME = $(pkgname)-$(pkgversion) INSTALL = install INSTALL_PROGRAM = $(INSTALL) -m 755 -INSTALL_DATA = $(INSTALL) -m 644 INSTALL_DIR = $(INSTALL) -d -m 755 +INSTALL_DATA = $(INSTALL) -m 644 SHELL = /bin/sh CAN_RUN_INSTALLINFO = $(SHELL) -c "install-info --version" > /dev/null 2>&1 @@ -150,11 +150,9 @@ dist : doc $(DISTNAME)/testsuite/test.txt \ $(DISTNAME)/testsuite/test21636.txt \ $(DISTNAME)/testsuite/test_bad[6-9].txt \ - $(DISTNAME)/testsuite/test_3m.txt.lz.md5 \ $(DISTNAME)/testsuite/fox.lz \ $(DISTNAME)/testsuite/fox_*.lz \ - $(DISTNAME)/testsuite/fox6.lz \ - $(DISTNAME)/testsuite/fox6_nz.lz \ + $(DISTNAME)/testsuite/fox6_b1nz.lz \ $(DISTNAME)/testsuite/fox6_sc[1-6].lz \ $(DISTNAME)/testsuite/fox6_bad[1-6].lz \ $(DISTNAME)/testsuite/numbers.lz \ @@ -162,7 +160,6 @@ dist : doc $(DISTNAME)/testsuite/test.txt.lz \ $(DISTNAME)/testsuite/test.txt.lzma \ $(DISTNAME)/testsuite/test_bad[1-9].lz \ - $(DISTNAME)/testsuite/test_em.txt.lz \ $(DISTNAME)/testsuite/test.txt.lz.fec \ $(DISTNAME)/testsuite/test.txt.lz.fec16 rm -f $(DISTNAME) @@ -12,20 +12,21 @@ The option '--fec-file', which sets the fec file to be used, has been added. The options '-r, --recursive' and '-R, --dereference-recursive' have been added for recursive creation and reading of fec files. -The short name of option '--byte-repair' has been changed to "-B". +The short name of option '--byte-repair' has been changed to '-B'. -The option '--ignore-empty', which makes lziprecover ignore empty members in -multimember files when decompressing, testing, or listing, has been added. -By default lziprecover now exits with error status 2 if any empty member is -found in a multimember file. +The option '--byte-repair' now repairs a nonzero first LZMA byte. -The option '--ignore-nonzero', which makes lziprecover ignore a nonzero -first byte in the LZMA stream when decompressing or testing, has been added. -By default lziprecover now exits with error status 2 if the first LZMA byte -is nonzero in any member of the input files. +When decompressing, testing, or listing, lziprecover now exits with error +status 2 if any empty member is found in a regular multimember file unless +'-i' is given. + +When decompressing or testing, lziprecover now exits with error status 2 if +the first byte of the LZMA stream is not 0 unless '-i' is given. The option '--clear-marking' has been renamed to '--nonzero-repair'. Options '--empty-error' and '--marking-error' have been removed. +The chapter 'Syntax of command-line arguments' has been added to the manual. + Lzip 1.16 (or clzip 1.6) or newer is required to run the tests. diff --git a/alone_to_lz.cc b/alone_to_lz.cc index c71f335..ff1f92f 100644 --- a/alone_to_lz.cc +++ b/alone_to_lz.cc @@ -52,7 +52,7 @@ uint8_t * read_file( const int infd, long * const file_sizep, if( buffer_size >= LONG_MAX ) { show_file_error( filename, large_file_msg ); std::free( buffer ); return 0; } - buffer_size = ( buffer_size <= LONG_MAX / 2 ) ? 2 * buffer_size : LONG_MAX; + buffer_size = (buffer_size <= LONG_MAX / 2) ? 2 * buffer_size : LONG_MAX; uint8_t * const tmp = (uint8_t *)std::realloc( buffer, buffer_size ); if( !tmp ) { std::free( buffer ); throw std::bad_alloc(); } buffer = tmp; @@ -143,8 +143,8 @@ int alone_to_lz( const int infd, const Pretty_print & pp ) { pp( "conversion failed" ); std::free( buffer ); return 2; } if( writeblock( outfd, buffer + offset, lzip_size ) != lzip_size ) { - show_error( "Error writing output file", errno ); - std::free( buffer ); return 1; + show_file_error( printable_name( output_filename, false ), write_error_msg, + errno ); std::free( buffer ); return 1; } std::free( buffer ); if( verbosity >= 1 ) std::fputs( "done\n", stderr ); diff --git a/byte_repair.cc b/byte_repair.cc index d6b1782..1591394 100644 --- a/byte_repair.cc +++ b/byte_repair.cc @@ -58,6 +58,19 @@ bool gross_damage( const uint8_t * const mbuffer, const long msize ) } +// Return value: 0 = errors remain, 6 = repaired pos +int repair_nonzero( uint8_t * const mbuffer, const long msize ) + { + mbuffer[6] = 0; + const Lzip_header & header = *(Lzip_header *)mbuffer; + const unsigned dictionary_size = header.dictionary_size(); + if( !isvalid_ds( dictionary_size ) ) return 0; + LZ_mtester mtester( mbuffer, msize, dictionary_size ); + if( mtester.test_member() == 0 ) return 6; + return 0; + } + + // Return value: 0 = no change, 5 = repaired pos int repair_dictionary_size( uint8_t * const mbuffer, const long msize ) { @@ -155,6 +168,15 @@ long repair_member( uint8_t * const mbuffer, const long long mpos, } // end namespace +bool safe_seek( const int fd, const long long pos, + const std::string & filename ) + { + if( lseek( fd, pos, SEEK_SET ) == pos ) return true; + show_file_error( filename.c_str(), "Seek error", errno ); + return false; + } + + long seek_write( const int fd, const uint8_t * const buf, const long size, const long long pos ) { @@ -165,16 +187,16 @@ long seek_write( const int fd, const uint8_t * const buf, const long size, uint8_t * read_member( const int infd, const long long mpos, - const long long msize, const char * const filename ) + const long long msize, const std::string & filename ) { if( msize <= 0 || msize > LONG_MAX ) - { show_file_error( filename, + { show_file_error( filename.c_str(), "Input file contains member larger than LONG_MAX." ); return 0; } if( !safe_seek( infd, mpos, filename ) ) return 0; uint8_t * const buffer = new uint8_t[msize]; if( readblock( infd, buffer, msize ) != msize ) - { show_file_error( filename, read_error_msg, errno ); + { show_file_error( filename.c_str(), read_error_msg, errno ); delete[] buffer; return 0; } return buffer; } @@ -206,8 +228,10 @@ int byte_repair( const std::string & input_filename, const long long msize = lzip_index.mblock( i ).size(); if( !safe_seek( infd, mpos, filename ) ) cleanup_and_fail( 1 ); long long failure_pos = 0; - if( test_member_from_file( infd, msize, &failure_pos ) == 0 ) continue; - if( failure_pos < Lzip_header::size ) // End Of File + bool nonzero = false; + const int ret = test_member_from_file( infd, msize, &failure_pos, &nonzero ); + if( ret == 0 && !nonzero ) continue; + if( ret != 0 && failure_pos < Lzip_header::size ) // End Of File { show_error( "Can't repair error in input file." ); cleanup_and_fail( 2 ); } if( failure_pos >= msize - 8 ) failure_pos = msize - 8 - 1; @@ -218,14 +242,17 @@ int byte_repair( const std::string & input_filename, i + 1, lzip_index.members(), mpos + failure_pos ); std::fflush( stdout ); } - uint8_t * const mbuffer = read_member( infd, mpos, msize, filename ); + uint8_t * const mbuffer = read_member( infd, mpos, msize, input_filename ); if( !mbuffer ) cleanup_and_fail( 1 ); const Lzip_header & header = *(const Lzip_header *)mbuffer; const unsigned dictionary_size = header.dictionary_size(); long pos = 0; + if( !nonzero && mbuffer[6] != 0 ) nonzero = true; // bad DS if( !gross_damage( mbuffer, msize ) ) { - pos = repair_dictionary_size( mbuffer, msize ); + if( nonzero ) pos = repair_nonzero( mbuffer, msize ); + if( pos == 0 ) + pos = repair_dictionary_size( mbuffer, msize ); if( pos == 0 ) pos = repair_member( mbuffer, mpos, msize, header.size + 1, header.size + 6, dictionary_size, terminator ); @@ -243,12 +270,14 @@ int byte_repair( const std::string & input_filename, if( !safe_seek( infd, 0, filename ) ) return 1; set_signal_handler(); if( !open_outstream( true, true, false, true, to_file ) ) return 1; - if( !copy_file( infd, outfd ) ) // copy whole file - cleanup_and_fail( 1 ); + if( !copy_file( infd, outfd, input_filename, output_filename ) ) + cleanup_and_fail( 1 ); // copy whole file } - if( seek_write( outfd, mbuffer + pos, 1, mpos + pos ) != 1 ) - { show_error( "Error writing output file", errno ); - cleanup_and_fail( 1 ); } + if( ( nonzero && pos != 6 && + seek_write( outfd, mbuffer + 6, 1, mpos + 6 ) != 1 ) || + seek_write( outfd, mbuffer + pos, 1, mpos + pos ) != 1 ) + { show_file_error( printable_name( output_filename, false ), + write_error_msg, errno ); cleanup_and_fail( 1 ); } } delete[] mbuffer; if( pos == 0 ) @@ -272,24 +301,24 @@ int byte_repair( const std::string & input_filename, } -int debug_delay( const char * const input_filename, +int debug_delay( const std::string & input_filename, const Cl_options & cl_opts, Block range, const char terminator ) { + const char * const filename = input_filename.c_str(); struct stat in_stats; // not used - const int infd = open_instream( input_filename, &in_stats, false, true ); + const int infd = open_instream( filename, &in_stats, false, true ); if( infd < 0 ) return 1; const Lzip_index lzip_index( infd, cl_opts ); if( lzip_index.retval() != 0 ) - { show_file_error( input_filename, lzip_index.error().c_str() ); + { show_file_error( filename, lzip_index.error().c_str() ); return lzip_index.retval(); } if( range.end() > lzip_index.cdata_size() ) range.size( std::max( 0LL, lzip_index.cdata_size() - range.pos() ) ); if( range.size() <= 0 ) - { show_file_error( input_filename, "Nothing to do; range is empty." ); - return 0; } + { show_file_error( filename, "Nothing to do; range is empty." ); return 0; } for( long i = 0; i < lzip_index.members(); ++i ) { @@ -355,24 +384,25 @@ int debug_delay( const char * const input_filename, } -int debug_byte_repair( const char * const input_filename, +int debug_byte_repair( const std::string & input_filename, const Cl_options & cl_opts, const Bad_byte & bad_byte, const char terminator ) { + const char * const filename = input_filename.c_str(); struct stat in_stats; // not used - const int infd = open_instream( input_filename, &in_stats, false, true ); + const int infd = open_instream( filename, &in_stats, false, true ); if( infd < 0 ) return 1; const Lzip_index lzip_index( infd, cl_opts ); if( lzip_index.retval() != 0 ) - { show_file_error( input_filename, lzip_index.error().c_str() ); + { show_file_error( filename, lzip_index.error().c_str() ); return lzip_index.retval(); } long idx = 0; for( ; idx < lzip_index.members(); ++idx ) if( lzip_index.mblock( idx ).includes( bad_byte.pos ) ) break; if( idx >= lzip_index.members() ) - { show_file_error( input_filename, "Nothing to do; byte is beyond EOF." ); + { show_file_error( filename, "Nothing to do; byte is beyond EOF." ); return 0; } const long long mpos = lzip_index.mblock( idx ).pos(); @@ -392,11 +422,12 @@ int debug_byte_repair( const char * const input_filename, if( !mbuffer ) return 1; const Lzip_header & header = *(const Lzip_header *)mbuffer; const unsigned dictionary_size = header.dictionary_size(); - const uint8_t good_value = mbuffer[bad_byte.pos-mpos]; + const long long bad_pos = bad_byte.pos - mpos; + const uint8_t good_value = mbuffer[bad_pos]; const uint8_t bad_value = bad_byte( good_value ); - mbuffer[bad_byte.pos-mpos] = bad_value; + mbuffer[bad_pos] = bad_value; long failure_pos = 0; - if( bad_byte.pos != 5 || isvalid_ds( header.dictionary_size() ) ) + if( bad_pos != 5 || isvalid_ds( header.dictionary_size() ) ) { LZ_mtester mtester( mbuffer, msize, header.dictionary_size() ); if( mtester.test_member() == 0 && mtester.finished() ) @@ -420,6 +451,8 @@ int debug_byte_repair( const char * const input_filename, if( failure_pos >= msize ) failure_pos = msize - 1; long pos = repair_dictionary_size( mbuffer, msize ); if( pos == 0 ) + if( mbuffer[6] != 0 ) pos = repair_nonzero( mbuffer, msize ); + if( pos == 0 ) pos = repair_member( mbuffer, mpos, msize, header.size + 1, header.size + 6, dictionary_size, terminator ); if( pos == 0 ) @@ -441,21 +474,21 @@ int debug_byte_repair( const char * const input_filename, (Packet sizes are a fractionary number of bytes. The packet and marker sizes shown by option -X are the number of extra bytes required to decode the packet, not counting the data present in the range decoder before and - after the decoding. The max marker size of a 'Sync Flush marker' does not - include the 5 bytes read by rdec.load). + after the decoding. if bad_byte.pos >= cdata_size, bad_byte is ignored. */ -int debug_decompress( const char * const input_filename, +int debug_decompress( const std::string & input_filename, const Cl_options & cl_opts, const Bad_byte & bad_byte, const bool show_packets ) { + const char * const filename = input_filename.c_str(); struct stat in_stats; - const int infd = open_instream( input_filename, &in_stats, false, true ); + const int infd = open_instream( filename, &in_stats, false, true ); if( infd < 0 ) return 1; const Lzip_index lzip_index( infd, cl_opts ); if( lzip_index.retval() != 0 ) - { show_file_error( input_filename, lzip_index.error().c_str() ); + { show_file_error( filename, lzip_index.error().c_str() ); return lzip_index.retval(); } outfd = show_packets ? -1 : STDOUT_FILENO; @@ -6,7 +6,7 @@ # to copy, distribute, and modify it. pkgname=lziprecover -pkgversion=1.25-pre1 +pkgversion=1.25-rc1 progname=lziprecover srctrigger=doc/${pkgname}.texi @@ -77,7 +77,7 @@ bool Range_decoder::read_block() { stream_pos = readblock( infd, buffer, buffer_size ); if( stream_pos != buffer_size && errno ) throw Error( read_error_msg ); - at_stream_end = ( stream_pos < buffer_size ); + at_stream_end = stream_pos < buffer_size; partial_member_pos += pos; pos = 0; show_dprogress(); @@ -99,7 +99,7 @@ void LZ_decoder::flush_data() const long long s = std::min( positive_diff( outend, sp ), (unsigned long long)size ) - i; if( s > 0 && writeblock( outfd, buffer + stream_pos + i, s ) != s ) - throw Error( "Write error" ); + throw Error( write_error_msg ); } if( pos >= dictionary_size ) { partial_data_pos += pos; pos = 0; pos_wrapped = true; } @@ -180,7 +180,8 @@ bool LZ_decoder::check_trailer( const Pretty_print & pp ) const /* Return value: 0 = OK, 1 = decoder error, 2 = unexpected EOF, 3 = trailer error, 4 = unknown marker found, 5 = nonzero first LZMA byte found. */ -int LZ_decoder::decode_member( const Pretty_print & pp, const bool ignore_nonzero ) +int LZ_decoder::decode_member( const Pretty_print & pp, + const bool ignore_nonzero ) { Bit_model bm_literal[1<<literal_context_bits][0x300]; Bit_model bm_match[State::states][pos_states]; @@ -244,22 +245,22 @@ int LZ_decoder::decode_member( const Pretty_print & pp, const bool ignore_nonzer } else // match { + rep3 = rep2; rep2 = rep1; rep1 = rep0; len = rdec.decode_len( match_len_model, pos_state ); - unsigned distance = rdec.decode_tree6( bm_dis_slot[get_len_state(len)] ); - if( distance >= start_dis_model ) + rep0 = rdec.decode_tree6( bm_dis_slot[get_len_state(len)] ); + if( rep0 >= start_dis_model ) { - const unsigned dis_slot = distance; + const unsigned dis_slot = rep0; const int direct_bits = ( dis_slot >> 1 ) - 1; - distance = ( 2 | ( dis_slot & 1 ) ) << direct_bits; + rep0 = ( 2 | ( dis_slot & 1 ) ) << direct_bits; if( dis_slot < end_dis_model ) - distance += rdec.decode_tree_reversed( - bm_dis + ( distance - dis_slot ), direct_bits ); + rep0 += rdec.decode_tree_reversed( bm_dis + ( rep0 - dis_slot ), + direct_bits ); else { - distance += - rdec.decode( direct_bits - dis_align_bits ) << dis_align_bits; - distance += rdec.decode_tree_reversed4( bm_align ); - if( distance == 0xFFFFFFFFU ) // marker found + rep0 += rdec.decode( direct_bits - dis_align_bits ) << dis_align_bits; + rep0 += rdec.decode_tree_reversed4( bm_align ); + if( rep0 == 0xFFFFFFFFU ) // marker found { rdec.normalize(); flush_data(); @@ -271,7 +272,6 @@ int LZ_decoder::decode_member( const Pretty_print & pp, const bool ignore_nonzer } } } - rep3 = rep2; rep2 = rep1; rep1 = rep0; rep0 = distance; state.set_match(); if( rep0 >= dictionary_size || ( rep0 >= pos && !pos_wrapped ) ) { flush_data(); return 1; } @@ -26,6 +26,7 @@ class Range_decoder uint32_t range; const int infd; // input file descriptor bool at_stream_end; + bool nonzero_; bool read_block(); @@ -42,11 +43,12 @@ public: code( 0 ), range( 0xFFFFFFFFU ), infd( ifd ), - at_stream_end( false ) + at_stream_end( false ), nonzero_( false ) {} ~Range_decoder() { delete[] buffer; } + bool nonzero() const { return nonzero_; } unsigned get_code() const { return code; } bool finished() { return pos >= stream_pos && !read_block(); } @@ -110,8 +112,10 @@ public: { code = 0; range = 0xFFFFFFFFU; - // check first byte of the LZMA stream - if( get_byte() != 0 && !ignore_nonzero ) return false; + // check first byte of the LZMA stream without reading it + nonzero_ = buffer[pos] != 0; + if( nonzero_ && !ignore_nonzero ) return false; + get_byte(); // discard first byte of the LZMA stream for( int i = 0; i < 4; ++i ) code = ( code << 8 ) | get_byte(); return true; } @@ -131,7 +135,7 @@ public: range >>= 1; // symbol <<= 1; // if( code >= range ) { code -= range; symbol |= 1; } - const bool bit = ( code >= range ); + const bool bit = code >= range; symbol <<= 1; symbol += bit; code -= range & ( 0U - bit ); } @@ -329,14 +333,14 @@ class LZ_decoder bool fast, fast2; if( lpos > distance ) { - fast = ( len < dictionary_size - lpos ); - fast2 = ( fast && len <= lpos - i ); + fast = len < dictionary_size - lpos; + fast2 = fast && len <= lpos - i; } else { i += dictionary_size; - fast = ( len < dictionary_size - i ); // (i == pos) may happen - fast2 = ( fast && len <= i - lpos ); + fast = len < dictionary_size - i; // (i == pos) may happen + fast2 = fast && len <= i - lpos; } if( fast ) // no wrap { diff --git a/doc/lziprecover.1 b/doc/lziprecover.1 index bb39f36..0fafa13 100644 --- a/doc/lziprecover.1 +++ b/doc/lziprecover.1 @@ -1,5 +1,5 @@ .\" DO NOT MODIFY THIS FILE! It was generated by help2man 1.49.2. -.TH LZIPRECOVER "1" "October 2024" "lziprecover 1.25-pre1" "User Commands" +.TH LZIPRECOVER "1" "November 2024" "lziprecover 1.25-rc1" "User Commands" .SH NAME lziprecover \- recovers data from damaged lzip files .SH SYNOPSIS @@ -119,12 +119,6 @@ remove members, tdata from files in place \fB\-\-strip=\fR<list>:d:e:t copy files to stdout stripping members given .TP -\fB\-\-ignore\-empty\fR -ignore empty members in multimember files -.TP -\fB\-\-ignore\-nonzero\fR -ignore a nonzero first LZMA byte -.TP \fB\-\-loose\-trailing\fR allow trailing data seeming corrupt header .TP diff --git a/doc/lziprecover.info b/doc/lziprecover.info index 197af5e..20fd014 100644 --- a/doc/lziprecover.info +++ b/doc/lziprecover.info @@ -12,12 +12,13 @@ File: lziprecover.info, Node: Top, Next: Introduction, Up: (dir) Lziprecover Manual ****************** -This manual is for Lziprecover (version 1.25-pre1, 1 October 2024). +This manual is for Lziprecover (version 1.25-rc1, 18 November 2024). * Menu: * Introduction:: Purpose and features of lziprecover * Invoking lziprecover:: Command-line interface +* Argument syntax:: By convention, options start with a hyphen * File format:: Detailed format of the compressed file * Data safety:: Protecting data from accidental loss * Fec files:: Forward Error Correction @@ -112,8 +113,9 @@ pdlzip. If the cause of file corruption is a damaged medium, the combination GNU ddrescue + lziprecover is the recommended option for recovering data -from damaged lzip files. *Note ddrescue-example::, and *note -ddrescue-example2::, for examples. +from damaged files. *Note ddrescue-example::, *note ddrescue-example2::, and +*note ddrescue-example3::, for examples. *Note GNU ddrescue manual: +(ddrescue)Top, for details about ddrescue. If a file is too damaged for lziprecover to repair it, all the recoverable data in all members of the file can be extracted with the @@ -135,7 +137,7 @@ have been compressed. Decompressed is used to refer to data which have undergone the process of decompression. -File: lziprecover.info, Node: Invoking lziprecover, Next: File format, Prev: Introduction, Up: Top +File: lziprecover.info, Node: Invoking lziprecover, Next: Argument syntax, Prev: Introduction, Up: Top 2 Invoking lziprecover ********************** @@ -150,8 +152,7 @@ first time it appears in the command line. If no file names are specified, lziprecover decompresses from standard input to standard output. Remember to prepend './' to any file name beginning with a hyphen, or use '--'. -lziprecover supports the following options: *Note Argument syntax: -(arg_parser)Argument syntax. +lziprecover supports the following options: *Note Argument syntax::. '-h' '--help' @@ -175,7 +176,7 @@ lziprecover supports the following options: *Note Argument syntax: dictionary size of the resulting file (and therefore the amount of memory required to decompress it). Only streamed files with default LZMA properties can be converted; non-streamed lzma-alone files lack - the "End Of Stream" marker required in lzip files. + the 'End Of Stream' marker required in lzip files. The name of the converted lzip file is derived from that of the original lzma-alone file as follows: @@ -215,23 +216,24 @@ lziprecover supports the following options: *Note Argument syntax: status 1. If a file fails to decompress, or is a terminal, lziprecover exits immediately with error status 2 without decompressing the rest of the files. A terminal is considered an uncompressed file, and - therefore invalid. + therefore invalid. A multimember file with one or more empty members + is accepted if redirected to standard input or if '-i' is given. '-D RANGE' '--range-decompress=RANGE' Decompress only a range of bytes starting at decompressed byte position - BEGIN and up to byte position END - 1. Byte positions start at 0. This - option provides random access to the data in multimember files; it - only decompresses the members containing the desired data. In order to - guarantee the correctness of the data produced, all members containing - any part of the desired data are decompressed and their integrity is - checked. + BEGIN and up to byte position END - 1. Byte positions start at 0. The + bytes produced are sent to standard output unless the option '-o' is + used. This option provides random access to the data in multimember + files; it only decompresses the members containing the desired data. + In order to guarantee the correctness of the data produced, all + members containing any part of the desired data are decompressed and + their integrity is checked. Four formats of RANGE are recognized, 'BEGIN', 'BEGIN-END', 'BEGIN,SIZE', and ',SIZE'. If only BEGIN is specified, END is taken as the end of the file. If only SIZE is specified, BEGIN is taken as the - beginning of the file. The bytes produced are sent to standard output - unless the option '--output' is used. + beginning of the file. '-e' '--reproduce' @@ -325,7 +327,8 @@ lziprecover supports the following options: *Note Argument syntax: '-k' '--keep' - Keep (don't delete) input files during decompression. + Keep (don't delete) input files during decompression or conversion from + lzma-alone. '-l' '--list' @@ -336,9 +339,11 @@ lziprecover supports the following options: *Note Argument syntax: '-v', the dictionary size, the number of members in the file, and the amount of trailing data (if any) are also printed. With '-vv', the positions and sizes of each member in multimember files are also - printed. With '-i', format errors are ignored, and with '-ivv', gaps - between members are shown. The member numbers shown coincide with the - file numbers produced by '--split'. + printed. A multimember file with one or more empty members is accepted + if redirected to standard input or if '-i' is given. With '-i', format + errors are ignored, and with '-ivv', gaps between members are shown. + The member numbers start at 1 and coincide with the file numbers + produced by '--split'. If any file is damaged, does not exist, can't be opened, or is not regular, the final exit status is > 0. '-lq' can be used to check @@ -358,8 +363,8 @@ lziprecover supports the following options: *Note Argument syntax: '-n N' '--threads=N' Set the maximum number of worker threads for '--fec=create', - overriding the system's default. Valid values range from 1 to "as many - as your system can support". If this option is not used, lziprecover + overriding the system's default. Valid values range from 1 to as many + as your system can support. If this option is not used, lziprecover tries to detect the number of processors in the system and use it as default value. 'lziprecover --help' shows the system's default value. @@ -367,7 +372,7 @@ lziprecover supports the following options: *Note Argument syntax: '--output=FILE[/]' If repairing, place the repaired output into FILE instead of into FILE_fixed.lz. If splitting, the names of the files produced are in - the form 'rec01FILE', 'rec02FILE', etc. + the form 'rec1FILE', 'rec2FILE', etc. If creating FEC data and '-c' has not been also specified, write the FEC data to FILE. If FILE ends with a slash, it is interpreted as the @@ -415,8 +420,8 @@ lziprecover supports the following options: *Note Argument syntax: headers or trailers, try to split FILE and then work on each member individually. - The names of the files produced are in the form 'rec01FILE', - 'rec02FILE', etc, and are designed so that the use of wildcards in + The names of the files produced are in the form 'rec1FILE', + 'rec2FILE', etc, and are designed so that the use of wildcards in subsequent processing, for example, 'lziprecover -cd rec*FILE > recovered_data', processes the files in the correct order. The number of digits used in the names varies @@ -430,7 +435,9 @@ lziprecover supports the following options: *Note Argument syntax: fails the test, does not exist, can't be opened, or is a terminal, lziprecover continues testing the rest of the files. A final diagnostic is shown at verbosity level 1 or higher if any file fails - the test when testing multiple files. + the test when testing multiple files. A multimember file with one or + more empty members is accepted if redirected to standard input or if + '-i' is given. '-v' '--verbose' @@ -448,14 +455,13 @@ lziprecover supports the following options: *Note Argument syntax: '--dump=[MEMBER_LIST][:damaged][:empty][:tdata]' Dump the members listed, the damaged members (if any), the empty members (if any), or the trailing data (if any) of one or more regular - multimember files to standard output, or to a file if the option - '--output' is used. If more than one file is given, the elements - dumped from all the files are concatenated. If a file does not exist, - can't be opened, or is not regular, lziprecover continues processing - the rest of the files. If the dump fails in one file, lziprecover - exits immediately without processing the rest of the files. Only - '--dump=tdata' can write to a terminal. '--dump=damaged' implies - '--ignore-errors'. + multimember files to standard output, or to a file if the option '-o' + is used. If more than one file is given, the elements dumped from all + the files are concatenated. If a file does not exist, can't be opened, + or is not regular, lziprecover continues processing the rest of the + files. If the dump fails in one file, lziprecover exits immediately + without processing the rest of the files. Only '--dump=tdata' can + write to a terminal. '--dump=damaged' implies '--ignore-errors'. The argument to '--dump' is a colon-separated list of the following element specifiers; a member list (1,3-6), a reverse member list @@ -509,35 +515,23 @@ lziprecover supports the following options: *Note Argument syntax: '--strip=[MEMBER_LIST][:damaged][:empty][:tdata]' Copy one or more regular multimember files to standard output (or to a - file if the option '--output' is used), stripping the members listed, - the damaged members (if any), the empty members (if any), or the - trailing data (if any) from each file. If all members in a file are - selected to be stripped, the trailing data (if any) are also stripped - even if 'tdata' is not specified. If more than one file is given, the - files are concatenated. In this case the trailing data are also - stripped from all but the last file even if 'tdata' is not specified. - If a file does not exist, can't be opened, or is not regular, - lziprecover continues processing the rest of the files. If a file - fails to copy, lziprecover exits immediately without processing the - rest of the files. See '--dump' above for a description of the - argument. - -'--ignore-empty' - When decompressing, testing, or listing, ignore empty members in - multimember files. By default lziprecover exits with error status 2 if - any empty member is found in a multimember file. - -'--ignore-nonzero' - When decompressing or testing, ignore a nonzero first byte in the LZMA - stream. By default lziprecover exits with error status 2 if the first - LZMA byte is nonzero in any member of the input files. Use - 'lziprecover --nonzero-repair' to repair any such nonzero bytes. + file if the option '-o' is used), stripping the members listed, the + damaged members (if any), the empty members (if any), or the trailing + data (if any) from each file. If all members in a file are selected to + be stripped, the trailing data (if any) are also stripped even if + 'tdata' is not specified. If more than one file is given, the files are + concatenated. In this case the trailing data are also stripped from + all but the last file even if 'tdata' is not specified. If a file does + not exist, can't be opened, or is not regular, lziprecover continues + processing the rest of the files. If a file fails to copy, lziprecover + exits immediately without processing the rest of the files. See + '--dump' above for a description of the argument. '--loose-trailing' When decompressing, testing, or listing, allow trailing data whose first bytes are so similar to the magic bytes of a lzip header that they can be confused with a corrupt header. Use this option if a file - triggers a "corrupt header" error and the cause is not indeed a + triggers a 'corrupt header' error and the cause is not indeed a corrupt header. '--nonzero-repair' @@ -625,14 +619,15 @@ lziprecover also supports the following debug options (for experts): Load the compressed FILE into memory, set the byte at POSITION to VALUE, and decompress the modified compressed data to standard output. If the damaged member can be decompressed to the end (just fails with - a CRC mismatch), the members following it are also decompressed. + a CRC mismatch), the members following it are also decompressed. *Note + --set-byte::, for a description of VALUE. '-X[POSITION,VALUE]' '--show-packets[=POSITION,VALUE]' Load the compressed FILE into memory, optionally set the byte at POSITION to VALUE, decompress the modified compressed data (discarding the output), and print to standard output descriptions of the LZMA - packets being decoded. + packets being decoded. *Note --set-byte::, for a description of VALUE. '-Y RANGE' '--debug-delay=RANGE' @@ -649,6 +644,7 @@ lziprecover also supports the following debug options (for experts): '--debug-byte-repair=POSITION,VALUE' Load the compressed FILE into memory, set the byte at POSITION to VALUE, and then try to repair the byte error. *Note --byte-repair::. + *Note --set-byte::, for a description of VALUE. '--gf16' Forces the use of GF(2^16) when creating FEC blocks even if the number @@ -681,9 +677,57 @@ corrupt or invalid input file, 3 for an internal consistency error (e.g., bug) which caused lziprecover to panic. -File: lziprecover.info, Node: File format, Next: Data safety, Prev: Invoking lziprecover, Up: Top +File: lziprecover.info, Node: Argument syntax, Next: File format, Prev: Invoking lziprecover, Up: Top + +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, unless the + argument is the empty string). 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. + +The syntax of options with an optional argument is +'-<short_option><argument>' (without whitespace), or +'--<long_option>=<argument>'. + + +File: lziprecover.info, Node: File format, Next: Data safety, Prev: Argument syntax, Up: Top -3 File format +4 File format ************* Perfection is reached, not when there is no longer anything to add, but @@ -737,7 +781,7 @@ not allowed in multimember files. Valid values for dictionary size range from 4 KiB to 512 MiB. 'LZMA stream' - The LZMA stream, finished by an "End Of Stream" marker. Uses default + The LZMA stream, terminated by an 'End Of Stream' marker. Uses default values for encoder properties. *Note Stream format: (lzip)Stream format, for a complete description. @@ -757,7 +801,7 @@ not allowed in multimember files. File: lziprecover.info, Node: Data safety, Next: Fec files, Prev: File format, Up: Top -4 Protecting data from accidental loss +5 Protecting data from accidental loss ************************************** It is a fact of life that sometimes data becomes corrupt. Software has @@ -803,7 +847,7 @@ with gzip and bzip2 with respect to data safety: File: lziprecover.info, Node: Merging with a backup, Next: Reproducing a mailbox, Up: Data safety -4.1 Recovering a file using a damaged backup +5.1 Recovering a file using a damaged backup ============================================ Let's suppose that you made a compressed backup of your valuable scientific @@ -830,7 +874,7 @@ possible to recover a file with thousands of errors. File: lziprecover.info, Node: Reproducing a mailbox, Prev: Merging with a backup, Up: Data safety -4.2 Recovering new messages using an old backup +5.2 Recovering new messages using an old backup =============================================== Let's suppose that you make periodic backups of your email messages stored @@ -876,15 +920,14 @@ identical backups (*note performance-of-merge::). File: lziprecover.info, Node: Fec files, Next: Repairing one byte, Prev: Data safety, Up: Top -5 Forward Error Correction +6 Forward Error Correction ************************** -"Forward Error Correction" (FEC) is any way of protecting data from -corruption by creating redundant data that can be used later to repair -errors in the protected data. Lziprecover uses a Hilbert-based Reed-Solomon -code to create one fec file (with extension '.fec') for each file that -needs to be protected. The fec files created by lziprecover are -reproducible. +Forward Error Correction (FEC) is any way of protecting data from corruption +by creating redundant data that can be used later to repair errors in the +protected data. Lziprecover uses a Hilbert-based Reed-Solomon code to create +one fec file (with extension '.fec') for each file that needs to be +protected. The fec files created by lziprecover are reproducible. Reed-Solomon is the most space-efficient Error Correcting Code (ECC) for data stored in block devices. It creates redundant FEC blocks in such a way @@ -892,8 +935,7 @@ that X FEC blocks allow the recuperation of any combination of up to X lost data blocks. All the blocks (data and FEC) are of the same size, which in fec files must be a multiple of 512 bytes. Reed-Solomon is not optimum for corruption affecting random single bits in a file because each corrupt bit -invalidates the whole block containing it. But in block devices, scattered -bit flips should not happen. +invalidates the whole block containing it. Usually, a corrupt file does not provide an indication of where the corruption is located. Therefore, each fec file stores one or two arrays of @@ -921,7 +963,7 @@ must be intact to provide 'prodata_size', 'prodata_md5', and 'gf16'. File: lziprecover.info, Node: How Reed-Solomon works, Next: Implementation details, Up: Fec files -5.1 How Reed-Solomon works +6.1 How Reed-Solomon works ========================== To illustrate how Reed-Solomon works on the BEC, we will use an example with @@ -944,8 +986,8 @@ p, q, and r can be computed from the values of x, y, and z: Now, if the values of x and y are lost because of data corruption, they can be recomputed by using any two of the three equations above. For -example, if we replace the known values of z, p, q, and r in equations (1) -and (2) we get: +example, if we replace the known values of z, p, and q in equations (1) and +(2) we get: x + y + 3 = 6 (1b) x + 2y + 9 = 14 (2b) @@ -982,7 +1024,7 @@ obtain the values of x and y (D = A^-1 * F): File: lziprecover.info, Node: Implementation details, Next: Creating fec files, Prev: How Reed-Solomon works, Up: Fec files -5.2 How lziprecover implements Reed-Solomon +6.2 How lziprecover implements Reed-Solomon =========================================== Lziprecover's implementation of Reed-Solomon can manage up to 128 data @@ -1011,17 +1053,17 @@ blocks. Lziprecover implements GF(2^8) with polynomial 0x11D and GF(2^16) with polynomial 0x1100B. - A Hilbert matrix is defined as 'A[i][j] = 1 / (i + j + 1)' for i and j ->= 0. But as in a Galois Field addition is exclusive or, applying the -Hilbert definition produces a singular (non invertible) matrix. To avoid -this problem, lziprecover uses a Hilbert matrix starting at row -'gf_size / 2'. I.e., 'A[i][j] = 1 / (i + gf_size / 2 + j)' for -'0 <= i,j < gf_size / 2'. (gf_size is the size of the Galois Field). + A Hilbert matrix is defined as A[i][j] = 1 / (i + j + 1) for i,j >= 0. +But, as in a Galois Field the addition is the exclusive or operation, +applying the Hilbert definition produces a singular (non invertible) +matrix. To avoid this problem, lziprecover uses a Hilbert matrix starting +at row r0 = gf_size / 2. I.e., A[i][j] = 1 / (i + j + r0) for +0 <= i,j < r0. ('gf_size' is the size of the Galois Field). File: lziprecover.info, Node: Creating fec files, Next: Testing with fec files, Prev: Implementation details, Up: Fec files -5.3 How to create fec files +6.3 How to create fec files =========================== Example 1: Create the fec file 'archive.tar.lz.fec' and store it in the @@ -1039,10 +1081,15 @@ Example 3: Create recursively one fec file for each file in the directory lziprecover -v -r -Fc -o fec/ datadir +Example 4: Create fec files for a collection of photos stored in directory +'photos' and store them in the directory 'photos-fec'. + + lziprecover -v -Fc -o photos-fec/ photos/* + File: lziprecover.info, Node: Testing with fec files, Next: Repairing with fec files, Prev: Creating fec files, Up: Fec files -5.4 How to test files using fec files +6.4 How to test files using fec files ===================================== Example 1: Test the integrity of 'archive.tar.lz' using the fec file @@ -1061,10 +1108,15 @@ directory 'fec'. lziprecover -v -r -Ft --fec-file=fec/ datadir +Example 4: Test the integrity of a collection of photos stored in directory +'photos' using fec files from directory 'photos-fec'. + + lziprecover -v -Ft --fec-file=photos-fec/ photos/* + File: lziprecover.info, Node: Repairing with fec files, Next: Fec file format, Prev: Testing with fec files, Up: Fec files -5.5 How to repair files using fec files +6.5 How to repair files using fec files ======================================= Example 1: Repair the file 'archive.tar.lz' using the fec file @@ -1084,10 +1136,22 @@ directory 'fec'. lziprecover -v -r -Fr --fec-file=fec/ datadir +Example 4: Recover a collection of photos from a damaged external drive +('/dev/sdc1'). The photos are in directory 'photos', and the fec files are +in directory 'photos-fec'. + + ddrescue -b4096 -r10 /dev/sdc1 hdimage mapfile + mount -o loop,ro hdimage /mnt/hdimage + cp -a /mnt/hdimage/photos photos + cp -a /mnt/hdimage/photos-fec photos-fec + umount /mnt/hdimage + lziprecover -v -Fr --fec-file=photos-fec/ photos/* + (Check and rename repaired files. They are named 'photos/*_fixed') + File: lziprecover.info, Node: Fec file format, Prev: Repairing with fec files, Up: Fec files -5.6 Fec file format +6.6 Fec file format =================== A fec file consists of one chksum packet, one or more fec packets, and one @@ -1127,7 +1191,7 @@ achieved by a careful design, without adding any padding bytes. The fec file format has an overhead of 8 bytes per protected data block, plus 16 bytes per FEC block, plus 80 bytes. -5.6.1 Chksum packet +6.6.1 Chksum packet ------------------- A chksum packet contains one CRC for each of the N data blocks in the @@ -1179,7 +1243,7 @@ payload_crc 36 + 4N 4 present) contains an array of CRC32-Cs. For the expected thousands of bit flips caused by a zeroed sector, a - "symmetric" CRC like CRC32 is probably better than CRC32-C, which + symmetric CRC like CRC32 is probably better than CRC32-C, which detects all the errors with an odd number of bit flips at the expense of a larger number of undetected errors with an even number of bit flips. @@ -1187,7 +1251,7 @@ payload_crc 36 + 4N 4 'payload_crc' CRC32 of the crc_array. -5.6.2 Fec packet +6.6.2 Fec packet ---------------- A fec packet contains one FEC block and is structured as shown in the @@ -1224,7 +1288,7 @@ payload_crc 12 + fbs 4 File: lziprecover.info, Node: Repairing one byte, Next: Merging files, Prev: Fec files, Up: Top -6 Repairing one byte +7 Repairing one byte ******************** Lziprecover can repair perfectly most files with small errors (up to one @@ -1238,11 +1302,11 @@ most common forms of data corruption. is limited to 2 GiB on 32-bit systems. The error may be located anywhere in the file except in the first 5 -bytes of each member header or in the 'Member size' field of the trailer -(last 8 bytes of each member). If the error is in the header it can be -easily repaired with a text editor like GNU Moe (*note File format::). If -the error is in the member size, it is enough to ignore the message about -'bad member size' when decompressing. +bytes of each member header (magic and version) or in the 'Member size' +field of the trailer (last 8 bytes of each member). If the error is in the +header it can be easily repaired with a text editor like GNU Moe (*note +File format::). If the error is in the member size, it is enough to ignore +the message about 'bad member size' when decompressing. Bit flip happens when one bit in the file is changed from 0 to 1 or vice versa. It may be caused by bad RAM or even by natural radiation. I have @@ -1252,7 +1316,7 @@ seen a case of bit flip in a file stored on an USB flash drive. transmission errors or I/O errors just affect one byte, or even one bit, of the file. Also, unlike magnetic media, where errors usually affect a whole sector, solid-state storage devices tend to produce single-byte errors, -making of lzip the perfect format for data stored on such devices. +which lziprecover can repair. Repairing a file can take some time. Small files or files with the error located near the beginning can be repaired in a few seconds. But repairing @@ -1266,7 +1330,7 @@ repairs more efficiently the worst errors. File: lziprecover.info, Node: Merging files, Next: Reproducing one sector, Prev: Repairing one byte, Up: Top -7 Merging files +8 Merging files *************** If you have several copies of a file but all of them are too damaged to @@ -1320,10 +1384,8 @@ identical to the original, in just 5 seconds: than the number of corrupt bytes (3104) because contiguous corrupt bytes are counted as a single multibyte error. - Example 1: Recover a compressed backup from two copies on CD-ROM with -error-checked merging of copies. *Note GNU ddrescue manual: (ddrescue)Top, -for details about ddrescue. +error-checked merging of copies. ddrescue -d -r1 -b2048 /dev/cdrom cdimage1 mapfile1 mount -t iso9660 -o loop,ro cdimage1 /mnt/cdimage @@ -1339,7 +1401,6 @@ for details about ddrescue. lziprecover -tv backup.tar.lz backup.tar.lz: ok - Example 2: Recover the first volume of those created with the command 'lzip -b 32MiB -S 650MB big_db' from two copies, 'big_db1_00001.lz' and 'big_db2_00001.lz', with member 07 damaged in the first copy, member 18 @@ -1354,7 +1415,7 @@ correct file produced is saved in 'big_db_00001.lz'. File: lziprecover.info, Node: Reproducing one sector, Next: Tarlz, Prev: Merging files, Up: Top -8 Reproducing one sector +9 Reproducing one sector ************************ Lziprecover can recover a zeroed sector in a lzip file by concatenating the @@ -1430,7 +1491,7 @@ header, and that the archive can be reproduced. The tarlz format has minimum overhead. It uses basic ustar headers, and only adds extended pax headers when they are required. -8.1 Performance of '--reproduce' +9.1 Performance of '--reproduce' ================================ Reproduce mode is especially useful when recovering a corrupt backup (or a @@ -1483,7 +1544,6 @@ for a different version of the software. Member reproduced successfully. Copy of input file reproduced successfully. - Example 2: Recover a damaged backup with a zeroed sector of 4096 bytes at file position 1019904, using as reference a previous backup. The damaged backup comes from a damaged partition copied with ddrescue. @@ -1505,7 +1565,6 @@ backup comes from a damaged partition copied with ddrescue. Member reproduced successfully. Copy of input file reproduced successfully. - Example 3: Recover a damaged backup with a zeroed sector of 4096 bytes at file position 1019904, using as reference a file from the filesystem. (If the zeroed sector encodes (part of) a tar header, the tarball can't be @@ -1541,8 +1600,8 @@ has been renamed. File: lziprecover.info, Node: Tarlz, Next: File names, Prev: Reproducing one sector, Up: Top -9 Options supporting the tar.lz format -************************************** +10 Options supporting the tar.lz format +*************************************** Tarlz is a massively parallel (multi-threaded) combined implementation of the tar archiver and the lzip compressor. @@ -1562,8 +1621,8 @@ alignment between tar members and lzip members minimizes the amount of data lost in case of corruption. In this chapter we'll explain the ways in which lziprecover can recover and process multimember tar.lz archives. -9.1 Recovering damaged multimember tar.lz archives -================================================== +10.1 Recovering damaged multimember tar.lz archives +=================================================== If you have several copies of the damaged archive, try merging them first because merging has a high probability of success. *Note Merging files::. If @@ -1604,8 +1663,8 @@ possible from each damaged member in 'bad_members.tar.lz': cd tmp tarlz --keep-damaged -xvf ../bad_members.tar.lz -9.2 Processing multimember tar.lz archives -========================================== +10.2 Processing multimember tar.lz archives +=========================================== Lziprecover is able to copy a list of members from a file to another. For example the command @@ -1618,7 +1677,7 @@ end-of-file blocks. File: lziprecover.info, Node: File names, Next: Trailing data, Prev: Tarlz, Up: Top -10 Names of the files produced by lziprecover +11 Names of the files produced by lziprecover ********************************************* The name of the fixed file produced by '--byte-repair' and '--merge' is @@ -1634,7 +1693,7 @@ string '_fixed' is inserted before the extension. File: lziprecover.info, Node: Trailing data, Next: Examples, Prev: File names, Up: Top -11 Extra data appended to the file +12 Extra data appended to the file ********************************** Sometimes extra data are found appended to a lzip file after the last @@ -1644,7 +1703,7 @@ member. Such trailing data may be: example when writing to a tape. It is safe to append any amount of padding zero bytes to a lzip file. - * Useful data added by the user; an "End Of File" string (to check that + * Useful data added by the user; an 'End Of File' string (to check that the file has not been truncated), a cryptographically secure hash, a description of file contents, etc. It is safe to append any amount of text to a lzip file as long as none of the first four bytes of the @@ -1691,7 +1750,6 @@ Example 1: Add a comment or description to a compressed file. # This command removes the comment from file.lz lziprecover --remove=tdata file.lz - Example 2: Add and check a cryptographically secure hash. (This may be convenient, but a separate copy of the hash must be kept in a safe place to guarantee that both file and hash have not been maliciously replaced). @@ -1703,7 +1761,7 @@ guarantee that both file and hash have not been maliciously replaced). File: lziprecover.info, Node: Examples, Next: Unzcrash, Prev: Trailing data, Up: Top -12 A small tutorial with examples +13 A small tutorial with examples ********************************* Example 1: Extract all the files from archive 'foo.tar.lz'. @@ -1763,7 +1821,7 @@ integrity of the resulting files. File: lziprecover.info, Node: Unzcrash, Next: Problems, Prev: Examples, Up: Top -13 Testing the robustness of decompressors +14 Testing the robustness of decompressors ****************************************** *Note --unzcrash::, for a faster way of testing the robustness of lzip. @@ -1849,10 +1907,11 @@ unzcrash supports the following options: '-B[SIZE][,VALUE]' '--block[=SIZE][,VALUE]' - Test block errors of given SIZE, simulating a whole sector I/O error. - SIZE defaults to 512 bytes. VALUE defaults to 0. By default, only - contiguous, non-overlapping blocks are tested, but this may be changed - with the option '--delta'. + Test block errors of given SIZE, simulating a whole sector I/O error + by setting all the bytes in the block to VALUE before attempting + decompression. SIZE defaults to 512 bytes. VALUE defaults to 0. By + default, only contiguous, non-overlapping blocks are tested, but this + may be changed with the option '--delta'. '-d N' '--delta=N' @@ -1918,7 +1977,7 @@ bug) which caused unzcrash to panic. File: lziprecover.info, Node: Problems, Next: Concept index, Prev: Unzcrash, Up: Top -14 Reporting bugs +15 Reporting bugs ***************** There are probably bugs in lziprecover. There are certainly errors and @@ -1939,6 +1998,7 @@ Concept index * Menu: +* argument syntax: Argument syntax. (line 6) * bugs: Problems. (line 6) * chksum packet: Fec file format. (line 46) * data safety: Data safety. (line 6) @@ -1973,40 +2033,43 @@ Concept index Tag Table: Node: Top226 -Node: Introduction1463 -Node: Invoking lziprecover6223 -Ref: --trailing-error7167 -Ref: --byte-repair8261 -Ref: range-format10138 -Ref: --reproduce10473 -Ref: --unzcrash28457 -Node: File format32896 -Node: Data safety35653 -Node: Merging with a backup37899 -Node: Reproducing a mailbox39162 -Node: Fec files41616 -Node: How Reed-Solomon works43945 -Node: Implementation details46119 -Node: Creating fec files48188 -Node: Testing with fec files48852 -Node: Repairing with fec files49619 -Node: Fec file format50437 -Ref: fbs53308 -Node: Repairing one byte55099 -Node: Merging files57208 -Ref: performance-of-merge58387 -Ref: ddrescue-example59996 -Node: Reproducing one sector61283 -Ref: performance-of-reproduce65220 -Ref: ddrescue-example267894 -Node: Tarlz70314 -Node: File names73981 -Node: Trailing data74714 -Node: Examples78028 -Ref: concat-example78600 -Node: Unzcrash79999 -Node: Problems86385 -Node: Concept index86937 +Node: Introduction1535 +Node: Invoking lziprecover6387 +Ref: --trailing-error7308 +Ref: --byte-repair8402 +Ref: range-format10483 +Ref: --reproduce10728 +Ref: --unzcrash28455 +Node: Argument syntax33048 +Node: File format35005 +Node: Data safety37759 +Node: Merging with a backup40005 +Node: Reproducing a mailbox41268 +Node: Fec files43722 +Node: How Reed-Solomon works45988 +Node: Implementation details48159 +Node: Creating fec files50224 +Node: Testing with fec files51068 +Node: Repairing with fec files52023 +Ref: ddrescue-example52841 +Node: Fec file format53351 +Ref: fbs56222 +Node: Repairing one byte58011 +Node: Merging files60103 +Ref: performance-of-merge61282 +Ref: ddrescue-example262890 +Node: Reproducing one sector64106 +Ref: performance-of-reproduce68043 +Ref: ddrescue-example370716 +Node: Tarlz73135 +Node: File names76808 +Node: Trailing data77541 +Node: Examples80854 +Ref: concat-example81426 +Node: Unzcrash82825 +Ref: --set-byte87437 +Node: Problems89295 +Node: Concept index89847 End Tag Table diff --git a/doc/lziprecover.texi b/doc/lziprecover.texi index 41f3641..98eeb8c 100644 --- a/doc/lziprecover.texi +++ b/doc/lziprecover.texi @@ -6,8 +6,8 @@ @finalout @c %**end of header -@set UPDATED 1 October 2024 -@set VERSION 1.25-pre1 +@set UPDATED 18 November 2024 +@set VERSION 1.25-rc1 @dircategory Compression @direntry @@ -38,6 +38,7 @@ This manual is for Lziprecover (version @value{VERSION}, @value{UPDATED}). @menu * Introduction:: Purpose and features of lziprecover * Invoking lziprecover:: Command-line interface +* Argument syntax:: By convention, options start with a hyphen * File format:: Detailed format of the compressed file * Data safety:: Protecting data from accidental loss * Fec files:: Forward Error Correction @@ -139,8 +140,16 @@ pdlzip. If the cause of file corruption is a damaged medium, the combination @w{GNU ddrescue + lziprecover} is the recommended option for recovering data -from damaged lzip files. @xref{ddrescue-example}, and -@ref{ddrescue-example2}, for examples. +from damaged files. @xref{ddrescue-example}, @ref{ddrescue-example2}, and +@ref{ddrescue-example3}, for examples. +@ifnothtml +@xref{Top,GNU ddrescue manual,,ddrescue}, +@end ifnothtml +@ifhtml +See the +@uref{http://www.gnu.org/software/ddrescue/manual/ddrescue_manual.html,,ddrescue manual} +@end ifhtml +for details about ddrescue. If a file is too damaged for lziprecover to repair it, all the recoverable data in all members of the file can be extracted with the following command @@ -186,11 +195,7 @@ standard output. Remember to prepend @file{./} to any file name beginning with a hyphen, or use @samp{--}. @noindent -lziprecover supports the following -@uref{http://www.nongnu.org/arg-parser/manual/arg_parser_manual.html#Argument-syntax,,options}: -@ifnothtml -@xref{Argument syntax,,,arg_parser}. -@end ifnothtml +lziprecover supports the following options: @xref{Argument syntax}. @table @code @item -h @@ -211,12 +216,12 @@ garbage that can be safely ignored. @xref{concat-example}. @item -A @itemx --alone-to-lz -Convert lzma-alone files to lzip format without recompressing, just -adding a lzip header and trailer. The conversion minimizes the -dictionary size of the resulting file (and therefore the amount of -memory required to decompress it). Only streamed files with default LZMA -properties can be converted; non-streamed lzma-alone files lack the "End -Of Stream" marker required in lzip files. +Convert lzma-alone files to lzip format without recompressing, just adding a +lzip header and trailer. The conversion minimizes the dictionary size of the +resulting file (and therefore the amount of memory required to decompress +it). Only streamed files with default LZMA properties can be converted; +non-streamed lzma-alone files lack the 'End Of Stream' marker required in +lzip files. The name of the converted lzip file is derived from that of the original lzma-alone file as follows: @@ -258,24 +263,27 @@ already exists and @option{--force} has not been specified, lziprecover continues decompressing the rest of the files and exits with error status 1. If a file fails to decompress, or is a terminal, lziprecover exits immediately with error status 2 without decompressing the rest of the files. -A terminal is considered an uncompressed file, and therefore invalid. +A terminal is considered an uncompressed file, and therefore invalid. A +multimember file with one or more empty members is accepted if redirected to +standard input or if '-i' is given. @item -D @var{range} @itemx --range-decompress=@var{range} Decompress only a range of bytes starting at decompressed byte position @var{begin} and up to byte position @w{@var{end} - 1}. Byte positions start -at 0. This option provides random access to the data in multimember files; -it only decompresses the members containing the desired data. In order to -guarantee the correctness of the data produced, all members containing any -part of the desired data are decompressed and their integrity is checked. +at 0. The bytes produced are sent to standard output unless the option +@option{-o} is used. This option provides random access to the data in +multimember files; it only decompresses the members containing the desired +data. In order to guarantee the correctness of the data produced, all +members containing any part of the desired data are decompressed and their +integrity is checked. @anchor{range-format} Four formats of @var{range} are recognized, @samp{@var{begin}}, @samp{@var{begin}-@var{end}}, @samp{@var{begin},@var{size}}, and @samp{,@var{size}}. If only @var{begin} is specified, @var{end} is taken as the end of the file. If only @var{size} is specified, @var{begin} is taken -as the beginning of the file. The bytes produced are sent to standard output -unless the option @option{--output} is used. +as the beginning of the file. @anchor{--reproduce} @item -e @@ -371,7 +379,8 @@ last) may be wrong. @item -k @itemx --keep -Keep (don't delete) input files during decompression. +Keep (don't delete) input files during decompression or conversion from +lzma-alone. @item -l @itemx --list @@ -381,9 +390,11 @@ even for multimember files. If more than one file is given, a final line containing the cumulative sizes is printed. With @option{-v}, the dictionary size, the number of members in the file, and the amount of trailing data (if any) are also printed. With @option{-vv}, the positions and sizes of each -member in multimember files are also printed. With @option{-i}, format errors -are ignored, and with @option{-ivv}, gaps between members are shown. The -member numbers shown coincide with the file numbers produced by @option{--split}. +member in multimember files are also printed. A multimember file with one or +more empty members is accepted if redirected to standard input or if '-i' is +given. With @option{-i}, format errors are ignored, and with @option{-ivv}, +gaps between members are shown. The member numbers start at 1 and coincide +with the file numbers produced by @option{--split}. If any file is damaged, does not exist, can't be opened, or is not regular, the final exit status is @w{> 0}. @option{-lq} can be used to check quickly @@ -402,8 +413,8 @@ the merge mode. @item -n @var{n} @itemx --threads=@var{n} Set the maximum number of worker threads for @option{--fec=create}, -overriding the system's default. Valid values range from 1 to "as many as -your system can support". If this option is not used, lziprecover tries to +overriding the system's default. Valid values range from 1 to as many as +your system can support. If this option is not used, lziprecover tries to detect the number of processors in the system and use it as default value. @w{@samp{lziprecover --help}} shows the system's default value. @@ -411,7 +422,7 @@ detect the number of processors in the system and use it as default value. @itemx --output=@var{file}[/] If repairing, place the repaired output into @var{file} instead of into @var{file}_fixed.lz. If splitting, the names of the files produced are in -the form @file{rec01@var{file}}, @file{rec02@var{file}}, etc. +the form @file{rec1@var{file}}, @file{rec2@var{file}}, etc. If creating FEC data and @option{-c} has not been also specified, write the FEC data to @var{file}. If @var{file} ends with a slash, it is interpreted @@ -458,8 +469,8 @@ members with corrupt headers or trailers. If other lziprecover functions fail to work on a multimember @var{file} because of damage in headers or trailers, try to split @var{file} and then work on each member individually. -The names of the files produced are in the form @file{rec01@var{file}}, -@file{rec02@var{file}}, etc, and are designed so that the use of wildcards +The names of the files produced are in the form @file{rec1@var{file}}, +@file{rec2@var{file}}, etc, and are designed so that the use of wildcards in subsequent processing, for example, @w{@samp{lziprecover -cd rec*@var{file} > recovered_data}}, processes the files in the correct order. The number of digits used in the names varies @@ -473,7 +484,8 @@ together with @option{-v} to see information about the files. If a file fails the test, does not exist, can't be opened, or is a terminal, lziprecover continues testing the rest of the files. A final diagnostic is shown at verbosity level 1 or higher if any file fails the test when testing multiple -files. +files. A multimember file with one or more empty members is accepted if +redirected to standard input or if '-i' is given. @item -v @itemx --verbose @@ -489,8 +501,8 @@ operations, and extra information (for example, the failed areas). @item --dump=[@var{member_list}][:damaged][:empty][:tdata] Dump the members listed, the damaged members (if any), the empty members (if any), or the trailing data (if any) of one or more regular multimember files -to standard output, or to a file if the option @option{--output} is used. If -more than one file is given, the elements dumped from all the files are +to standard output, or to a file if the option @option{-o} is used. If more +than one file is given, the elements dumped from all the files are concatenated. If a file does not exist, can't be opened, or is not regular, lziprecover continues processing the rest of the files. If the dump fails in one file, lziprecover exits immediately without processing the rest of the @@ -547,7 +559,7 @@ attempting the removal of trailing data. @item --strip=[@var{member_list}][:damaged][:empty][:tdata] Copy one or more regular multimember files to standard output (or to a file -if the option @option{--output} is used), stripping the members listed, the +if the option @option{-o} is used), stripping the members listed, the damaged members (if any), the empty members (if any), or the trailing data (if any) from each file. If all members in a file are selected to be stripped, the trailing data (if any) are also stripped even if @samp{tdata} @@ -559,22 +571,11 @@ the rest of the files. If a file fails to copy, lziprecover exits immediately without processing the rest of the files. See @option{--dump} above for a description of the argument. -@item --ignore-empty -When decompressing, testing, or listing, ignore empty members in multimember -files. By default lziprecover exits with error status 2 if any empty member -is found in a multimember file. - -@item --ignore-nonzero -When decompressing or testing, ignore a nonzero first byte in the LZMA -stream. By default lziprecover exits with error status 2 if the first LZMA -byte is nonzero in any member of the input files. -Use @w{@samp{lziprecover --nonzero-repair}} to repair any such nonzero bytes. - @item --loose-trailing When decompressing, testing, or listing, allow trailing data whose first bytes are so similar to the magic bytes of a lzip header that they can be confused with a corrupt header. Use this option if a file triggers a -"corrupt header" error and the cause is not indeed a corrupt header. +'corrupt header' error and the cause is not indeed a corrupt header. @item --nonzero-repair Repair in place a nonzero first LZMA byte in the files specified. With @@ -666,13 +667,14 @@ Load the compressed @var{file} into memory, set the byte at @var{position} to @var{value}, and decompress the modified compressed data to standard output. If the damaged member can be decompressed to the end (just fails with a CRC mismatch), the members following it are also decompressed. +@xref{--set-byte}, for a description of @var{value}. @item -X[@var{position},@var{value}] @itemx --show-packets[=@var{position},@var{value}] Load the compressed @var{file} into memory, optionally set the byte at @var{position} to @var{value}, decompress the modified compressed data (discarding the output), and print to standard output descriptions of the -LZMA packets being decoded. +LZMA packets being decoded. @xref{--set-byte}, for a description of @var{value}. @item -Y @var{range} @itemx --debug-delay=@var{range} @@ -689,6 +691,7 @@ description of @var{range}. @itemx --debug-byte-repair=@var{position},@var{value} Load the compressed @var{file} into memory, set the byte at @var{position} to @var{value}, and then try to repair the byte error. @xref{--byte-repair}. +@xref{--set-byte}, for a description of @var{value}. @item --gf16 Forces the use of GF(2^16) when creating FEC blocks even if the number of @@ -723,6 +726,59 @@ indicate a corrupt or invalid input file, 3 for an internal consistency error (e.g., bug) which caused lziprecover 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, unless the +argument is the empty string). +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 + +@noindent +The syntax of options with an optional argument is +@option{-<short_option><argument>} (without whitespace), or +@option{--<long_option>=<argument>}. + + @node File format @chapter File format @cindex file format @@ -785,7 +841,7 @@ Example: 0xD3 = 2^19 - 6 * 2^15 = 512 KiB - 6 * 32 KiB = 320 KiB@* Valid values for dictionary size range from 4 KiB to 512 MiB. @item LZMA stream -The LZMA stream, finished by an "End Of Stream" marker. Uses default values +The LZMA stream, terminated by an 'End Of Stream' marker. Uses default values for encoder properties. @ifnothtml @xref{Stream format,,,lzip}, @@ -932,12 +988,11 @@ identical backups (@pxref{performance-of-merge}). @chapter Forward Error Correction @cindex forward error correction -"Forward Error Correction" (FEC) is any way of protecting data from -corruption by creating redundant data that can be used later to repair -errors in the protected data. Lziprecover uses a Hilbert-based Reed-Solomon -code to create one fec file (with extension @file{.fec}) for each file that -needs to be protected. The fec files created by lziprecover are -reproducible. +Forward Error Correction (FEC) is any way of protecting data from corruption +by creating redundant data that can be used later to repair errors in the +protected data. Lziprecover uses a Hilbert-based Reed-Solomon code to create +one fec file (with extension @file{.fec}) for each file that needs to be +protected. The fec files created by lziprecover are reproducible. Reed-Solomon is the most space-efficient Error Correcting Code (ECC) for data stored in block devices. It creates redundant FEC blocks in such a way @@ -945,8 +1000,7 @@ that X FEC blocks allow the recuperation of any combination of up to X lost data blocks. All the blocks (data and FEC) are of the same size, which in fec files must be a multiple of 512 bytes. Reed-Solomon is not optimum for corruption affecting random single bits in a file because each corrupt bit -invalidates the whole block containing it. But in block devices, scattered -bit flips should not happen. +invalidates the whole block containing it. Usually, a corrupt file does not provide an indication of where the corruption is located. Therefore, each fec file stores one or two arrays of @@ -1000,8 +1054,7 @@ If we have that x = 1, y = 2, and z = 3, then p = 6, q = 14, and r = 13: Now, if the values of x and y are lost because of data corruption, they can be recomputed by using any two of the three equations above. For example, if -we replace the known values of z, p, q, and r in equations (1) and (2) we -get: +we replace the known values of z, p, and q in equations (1) and (2) we get: @example x + y + 3 = 6 (1b) @@ -1076,13 +1129,12 @@ missing data blocks. Lziprecover implements GF(2^8) with polynomial 0x11D and GF(2^16) with polynomial 0x1100B. -A Hilbert matrix is defined as @w{@samp{A[i][j] = 1 / (i + j + 1)}} for i -and j >= 0. But as in a Galois Field addition is exclusive or, applying the -Hilbert definition produces a singular (non invertible) matrix. To avoid -this problem, lziprecover uses a Hilbert matrix starting at row -@w{@samp{gf_size / 2}}. I.e., @w{@samp{A[i][j] = 1 / (i + gf_size / 2 + j)}} -for @w{@samp{0 <= i,j < gf_size / 2}}. (gf_size is the size of the Galois -Field). +A Hilbert matrix is defined as @w{A[i][j] = 1 / (i + j + 1)} for +@w{i,j >= 0}. But, as in a Galois Field the addition is the exclusive or +operation, applying the Hilbert definition produces a singular (non +invertible) matrix. To avoid this problem, lziprecover uses a Hilbert matrix +starting at row @w{r0 = gf_size / 2}. I.e., @w{A[i][j] = 1 / (i + j + r0)} +for @w{0 <= i,j < r0}. (@samp{gf_size} is the size of the Galois Field). @node Creating fec files @@ -1113,6 +1165,14 @@ Example 3: Create recursively one fec file for each file in the directory lziprecover -v -r -Fc -o fec/ datadir @end example +@noindent +Example 4: Create fec files for a collection of photos stored in directory +@file{photos} and store them in the directory @file{photos-fec}. + +@example +lziprecover -v -Fc -o photos-fec/ photos/* +@end example + @node Testing with fec files @section How to test files using fec files @@ -1143,6 +1203,14 @@ directory @file{fec}. lziprecover -v -r -Ft --fec-file=fec/ datadir @end example +@noindent +Example 4: Test the integrity of a collection of photos stored in directory +@file{photos} using fec files from directory @file{photos-fec}. + +@example +lziprecover -v -Ft --fec-file=photos-fec/ photos/* +@end example + @node Repairing with fec files @section How to repair files using fec files @@ -1174,6 +1242,22 @@ directory @file{fec}. lziprecover -v -r -Fr --fec-file=fec/ datadir @end example +@anchor{ddrescue-example} +@noindent +Example 4: Recover a collection of photos from a damaged external drive +(@file{/dev/sdc1}). The photos are in directory @file{photos}, and the fec +files are in directory @file{photos-fec}. + +@example +ddrescue -b4096 -r10 /dev/sdc1 hdimage mapfile +mount -o loop,ro hdimage /mnt/hdimage +cp -a /mnt/hdimage/photos photos +cp -a /mnt/hdimage/photos-fec photos-fec +umount /mnt/hdimage +lziprecover -v -Fr --fec-file=photos-fec/ photos/* + (Check and rename repaired files. They are named @file{photos/*_fixed}) +@end example + @node Fec file format @section Fec file format @@ -1274,9 +1358,9 @@ The first chksum packet contains an array of CRC32s, while the second chksum packet (if present) contains an array of CRC32-Cs. For the expected thousands of bit flips caused by a zeroed sector, a -"symmetric" CRC like CRC32 is probably better than CRC32-C, which detects -all the errors with an odd number of bit flips at the expense of a larger -number of undetected errors with an even number of bit flips. +symmetric CRC like CRC32 is probably better than CRC32-C, which detects all +the errors with an odd number of bit flips at the expense of a larger number +of undetected errors with an even number of bit flips. @item payload_crc CRC32 of the crc_array. @@ -1334,9 +1418,9 @@ The file is repaired in memory. Therefore, enough virtual memory @w{(RAM + swap)} to contain the largest damaged member is required. Member size is limited to @w{2 GiB} on 32-bit systems. -The error may be located anywhere in the file except in the first 5 -bytes of each member header or in the @samp{Member size} field of the -trailer (last 8 bytes of each member). If the error is in the header it +The error may be located anywhere in the file except in the first 5 bytes of +each member header (magic and version) or in the @samp{Member size} field of +the trailer (last 8 bytes of each member). If the error is in the header it can be easily repaired with a text editor like GNU Moe (@pxref{File format}). If the error is in the member size, it is enough to ignore the message about @samp{bad member size} when decompressing. @@ -1349,7 +1433,7 @@ One byte may seem small, but most file corruptions not produced by transmission errors or I/O errors just affect one byte, or even one bit, of the file. Also, unlike magnetic media, where errors usually affect a whole sector, solid-state storage devices tend to produce single-byte -errors, making of lzip the perfect format for data stored on such devices. +errors, which lziprecover can repair. Repairing a file can take some time. Small files or files with the error located near the beginning can be repaired in a few seconds. But @@ -1421,19 +1505,10 @@ Note that the number of errors reported by lziprecover (2552) is lower than the number of corrupt bytes (3104) because contiguous corrupt bytes are counted as a single multibyte error. -@sp 1 -@anchor{ddrescue-example} +@anchor{ddrescue-example2} @noindent Example 1: Recover a compressed backup from two copies on CD-ROM with error-checked merging of copies. -@ifnothtml -@xref{Top,GNU ddrescue manual,,ddrescue}, -@end ifnothtml -@ifhtml -See the -@uref{http://www.gnu.org/software/ddrescue/manual/ddrescue_manual.html,,ddrescue manual} -@end ifhtml -for details about ddrescue. @example ddrescue -d -r1 -b2048 /dev/cdrom cdimage1 mapfile1 @@ -1451,7 +1526,6 @@ lziprecover -tv backup.tar.lz backup.tar.lz: ok @end example -@sp 1 @noindent Example 2: Recover the first volume of those created with the command @w{@samp{lzip -b 32MiB -S 650MB big_db}} from two copies, @@ -1608,8 +1682,7 @@ Member reproduced successfully. Copy of input file reproduced successfully. @end example -@sp 1 -@anchor{ddrescue-example2} +@anchor{ddrescue-example3} @noindent Example 2: Recover a damaged backup with a zeroed sector of 4096 bytes at file position 1019904, using as reference a previous backup. The damaged @@ -1634,7 +1707,6 @@ Member reproduced successfully. Copy of input file reproduced successfully. @end example -@sp 1 @noindent Example 3: Recover a damaged backup with a zeroed sector of 4096 bytes at file position 1019904, using as reference a file from the filesystem. (If @@ -1790,7 +1862,7 @@ example when writing to a tape. It is safe to append any amount of padding zero bytes to a lzip file. @item -Useful data added by the user; an "End Of File" string (to check that the +Useful data added by the user; an 'End Of File' string (to check that the file has not been truncated), a cryptographically secure hash, a description of file contents, etc. It is safe to append any amount of text to a lzip file as long as none of the first four bytes of the text matches the @@ -1844,7 +1916,6 @@ lziprecover --strip=tdata file.lz > stripped_file.lz lziprecover --remove=tdata file.lz @end example -@sp 1 @noindent Example 2: Add and check a cryptographically secure hash. (This may be convenient, but a separate copy of the hash must be kept in a safe place @@ -2036,10 +2107,11 @@ The number of N-bit errors per byte (N = 1 to 8) is: @item -B[@var{size}][,@var{value}] @itemx --block[=@var{size}][,@var{value}] -Test block errors of given @var{size}, simulating a whole sector I/O error. -@var{size} defaults to 512 bytes. @var{value} defaults to 0. By default, -only contiguous, non-overlapping blocks are tested, but this may be changed -with the option @option{--delta}. +Test block errors of given @var{size}, simulating a whole sector I/O error +by setting all the bytes in the block to @var{value} before attempting +decompression. @var{size} defaults to 512 bytes. @var{value} defaults to 0. +By default, only contiguous, non-overlapping blocks are tested, but this may +be changed with the option @option{--delta}. @item -d @var{n} @itemx --delta=@var{n} @@ -2049,6 +2121,7 @@ non-overlapping blocks, or truncation sizes. Values of @var{n} smaller than the block size result in overlapping blocks. (Which is convenient for testing because there are usually too few non-overlapping blocks in a file). +@anchor{--set-byte} @item -e @var{position},@var{value} @itemx --set-byte=@var{position},@var{value} Set byte at @var{position} to @var{value} in the internal buffer after diff --git a/dump_remove.cc b/dump_remove.cc index 4b90a92..a1a6757 100644 --- a/dump_remove.cc +++ b/dump_remove.cc @@ -59,7 +59,7 @@ int dump_members( const std::vector< std::string > & filenames, bool stdin_used = false; for( unsigned i = 0; i < filenames.size(); ++i ) { - const bool from_stdin = ( filenames[i] == "-" ); + const bool from_stdin = filenames[i] == "-"; if( from_stdin ) { if( stdin_used ) continue; else stdin_used = true; } const char * const input_filename = from_stdin ? "(stdin)" : filenames[i].c_str(); @@ -93,8 +93,8 @@ int dump_members( const std::vector< std::string > & filenames, if( in == !strip ) { if( !safe_seek( infd, stream_pos, input_filename ) || - !copy_file( infd, outfd, mb.pos() - stream_pos ) ) - cleanup_and_fail( 1 ); + !copy_file( infd, outfd, filenames[i], output_filename, + mb.pos() - stream_pos ) ) cleanup_and_fail( 1 ); copied_size += mb.pos() - stream_pos; ++members; } else { stripped_size += mb.pos() - stream_pos; ++smembers; } @@ -106,12 +106,13 @@ int dump_members( const std::vector< std::string > & filenames, if( !in && member_list.damaged ) { if( !safe_seek( infd, mb.pos(), input_filename ) ) cleanup_and_fail( 1 ); - in = ( test_member_from_file( infd, mb.size() ) != 0 ); // damaged + in = test_member_from_file( infd, mb.size() ) != 0; // damaged } if( in == !strip ) { if( !safe_seek( infd, mb.pos(), input_filename ) || - !copy_file( infd, outfd, mb.size() ) ) cleanup_and_fail( 1 ); + !copy_file( infd, outfd, filenames[i], output_filename, + mb.size() ) ) cleanup_and_fail( 1 ); copied_size += mb.size(); ++members; } else { stripped_size += mb.size(); ++smembers; } @@ -131,7 +132,8 @@ int dump_members( const std::vector< std::string > & filenames, ( !strip || i + 1 >= filenames.size() ) ) // strip all but last { if( !safe_seek( infd, cdata_size, input_filename ) || - !copy_file( infd, outfd, trailing_size ) ) cleanup_and_fail( 1 ); + !copy_file( infd, outfd, filenames[i], output_filename, + trailing_size ) ) cleanup_and_fail( 1 ); copied_tsize += trailing_size; } else if( trailing_size > 0 ) { stripped_tsize += trailing_size; ++tfiles; } @@ -210,7 +212,8 @@ int remove_members( const std::vector< std::string > & filenames, if( stream_pos != prev_end && ( !safe_seek( infd, prev_end, filename ) || !safe_seek( fd, stream_pos, filename ) || - !copy_file( infd, fd, mb.pos() - prev_end ) ) ) + !copy_file( infd, fd, filenames[i], filenames[i], + mb.pos() - prev_end ) ) ) { error = true; set_retval( retval, 1 ); break; } stream_pos += mb.pos() - prev_end; } @@ -224,14 +227,14 @@ int remove_members( const std::vector< std::string > & filenames, { if( !safe_seek( infd, mb.pos(), filename ) ) { error = true; set_retval( retval, 1 ); break; } - in = ( test_member_from_file( infd, mb.size() ) != 0 ); // damaged + in = test_member_from_file( infd, mb.size() ) != 0; // damaged } if( !in ) { if( stream_pos != mb.pos() && ( !safe_seek( infd, mb.pos(), filename ) || !safe_seek( fd, stream_pos, filename ) || - !copy_file( infd, fd, mb.size() ) ) ) + !copy_file( infd, fd, filenames[i], filenames[i], mb.size() ) ) ) { error = true; set_retval( retval, 1 ); break; } stream_pos += mb.size(); } @@ -254,7 +257,7 @@ int remove_members( const std::vector< std::string > & filenames, if( stream_pos != cdata_size && ( !safe_seek( infd, cdata_size, filename ) || !safe_seek( fd, stream_pos, filename ) || - !copy_file( infd, fd, trailing_size ) ) ) + !copy_file( infd, fd, filenames[i], filenames[i], trailing_size ) ) ) { close( fd ); close( infd ); set_retval( retval, 1 ); break; } stream_pos += trailing_size; } @@ -310,8 +313,7 @@ int nonzero_repair( const std::vector< std::string > & filenames, const int fd = open_truncable_stream( filename, &in_stats ); if( fd < 0 ) { set_retval( retval, 1 ); continue; } - const Lzip_index lzip_index( fd, cl_opts, cl_opts.ignore_errors, - cl_opts.ignore_errors ); + const Lzip_index lzip_index( fd, cl_opts, true, cl_opts.ignore_errors ); if( lzip_index.retval() != 0 ) { show_file_error( filename, lzip_index.error().c_str() ); @@ -332,7 +334,7 @@ int nonzero_repair( const std::vector< std::string > & filenames, if( seek_read( fd, header_buf, bufsize, mb.pos() ) != bufsize ) { show_file_error( filename, "Error reading member header", errno ); set_retval( retval, 1 ); break; } - if( !header.check( cl_opts.ignore_errors ) ) + if( !header.check( true ) ) { show_file_error( filename, "Member header became corrupt as we read it." ); set_retval( retval, 2 ); break; } if( *mark == 0 ) continue; @@ -234,7 +234,6 @@ void extract_dirname( const std::string & name, std::string & srcdir ); void replace_dirname( const std::string & name, const std::string & srcdir, const std::string & destdir, std::string & outname ); bool has_fec_extension( const std::string & name ); -const char * printable_name( const std::string & filename, const bool in = true ); int fec_create( const std::vector< std::string > & filenames, const std::string & default_output_filename, const unsigned long fb_or_pct, const unsigned cl_block_size, diff --git a/fec_create.cc b/fec_create.cc index e5c2a6c..90b79e1 100644 --- a/fec_create.cc +++ b/fec_create.cc @@ -241,7 +241,7 @@ Coded_fbs compute_fbs( const unsigned long prodata_size, { const unsigned unit_fbs = isvalid_fbs( cl_block_size ) ? cl_block_size : compute_unit_fbs( prodata_size ); - const unsigned long max_k = ( fec_level == 0 ) ? max_k8 : max_k16; + const unsigned long max_k = (fec_level == 0) ? max_k8 : max_k16; const unsigned k9 = std::min( ceil_divide( prodata_size, unit_fbs ), max_k ); const unsigned long fbsu9 = divide_fbs( prodata_size, k9, unit_fbs ); const unsigned long fbsu0 = divide_fbs( prodata_size, max_k8, unit_fbs ); @@ -258,9 +258,9 @@ unsigned compute_fec_blocks( const unsigned long prodata_size, { const unsigned long fbs = coded_fbs.val(); const unsigned prodata_blocks = ceil_divide( prodata_size, fbs ); - const unsigned long max_k = ( fec_level == 0 ) ? max_k8 : max_k16; + const unsigned long max_k = (fec_level == 0) ? max_k8 : max_k16; if( !isvalid_fbs( fbs ) || prodata_blocks > max_k ) return 0; - const unsigned long max_nk = ( fec_level == 0 ) ? max_k8 : max_nk16; + const unsigned long max_nk = (fec_level == 0) ? max_k8 : max_nk16; unsigned fec_blocks; if( fctype == fc_blocks ) fec_blocks = std::min( max_nk, fb_or_pct ); else @@ -387,7 +387,8 @@ bool write_fec( const char * const input_filename, return true; } fail: - show_file_error( input_filename, "Write error", errno ); return false; + show_file_error( printable_name( output_filename, false ), write_error_msg, + errno ); return false; } @@ -523,13 +524,6 @@ bool has_fec_extension( const std::string & name ) } -const char * printable_name( const std::string & filename, const bool in ) - { - if( filename.empty() || filename == "-" ) return in ? "(stdin)" : "(stdout)"; - return filename.c_str(); - } - - int fec_create( const std::vector< std::string > & filenames, const std::string & default_output_filename, const unsigned long fb_or_pct, const unsigned cl_block_size, diff --git a/fec_repair.cc b/fec_repair.cc index c15dfd3..8d9cd5b 100644 --- a/fec_repair.cc +++ b/fec_repair.cc @@ -76,7 +76,7 @@ uint8_t * read_file( const std::string & filename, long * const file_sizep ) { struct stat in_stats; // not used const char * const filenamep = printable_name( filename ); - const int infd = ( filename == "-" ) ? + const int infd = (filename == "-") ? STDIN_FILENO : open_instream( filenamep, &in_stats, false ); if( infd < 0 ) return 0; long buffer_size = 65536; @@ -88,7 +88,7 @@ uint8_t * read_file( const std::string & filename, long * const file_sizep ) if( buffer_size >= LONG_MAX ) { show_file_error( filenamep, large_file_msg ); std::free( buffer ); return 0; } - buffer_size = ( buffer_size <= LONG_MAX / 2 ) ? 2 * buffer_size : LONG_MAX; + buffer_size = (buffer_size <= LONG_MAX / 2) ? 2 * buffer_size : LONG_MAX; uint8_t * const tmp = (uint8_t *)std::realloc( buffer, buffer_size ); if( !tmp ) { show_file_error( filenamep, mem_msg ); std::free( buffer ); return 0; } @@ -794,9 +794,12 @@ int fec_test( const std::vector< std::string > & filenames, // check tty only once and don't try to delete a tty if( !open_outstream( force, false ) || !check_tty_out() ) return 1; } - if( writeblock( outfd, prodata, prodata_size ) != prodata_size || - !close_outstream( &in_stats ) ) // write repaired prodata - { set_retval( retval, 1 ); cleanup_and_fail( retval ); } + // write repaired prodata + if( writeblock( outfd, prodata, prodata_size ) != prodata_size ) + { show_file_error( printable_name( output_filename, false ), + write_error_msg, errno ); set_retval( retval, 1 ); } + else if( !close_outstream( &in_stats ) ) set_retval( retval, 1 ); + if( retval ) cleanup_and_fail( retval ); if( verbosity >= 1 ) std::fprintf( stderr, "Repaired copy of '%s' written to '%s'\n", input_filenamep, printable_name( output_filename, false ) ); @@ -868,7 +871,7 @@ int fec_dc( const std::string & input_filename, const unsigned long cblock_size = fec_blocks / cblocks * fbs; const unsigned long max_saved_size = cblocks * cblock_size; uint8_t * const sbuf = new uint8_t[max_saved_size]; // saved data bytes - const long last_pos = ( prodata_size % cblock_size != 0 ) ? + const long last_pos = (prodata_size % cblock_size != 0) ? prodata_size - prodata_size % cblock_size : prodata_size - cblock_size; if( verbosity >= 0 ) { std::printf( "Testing sets of %u block%s of size %s\n", cblocks, @@ -1034,7 +1037,7 @@ int fec_dZ( const std::string & input_filename, const int rest = std::min( 2UL, sector_size % fbs ); const long max_saved_size = ( sector_size / fbs + rest ) * fbs; uint8_t * const sbuf = new uint8_t[max_saved_size]; // saved data bytes - const long last_pos = ( prodata_size % sector_size != 0 ) ? + const long last_pos = (prodata_size % sector_size != 0) ? prodata_size - prodata_size % sector_size : prodata_size - sector_size; if( verbosity >= 0 ) { std::printf( "Testing blocks of size %s (delta %s)\n", @@ -1042,8 +1045,8 @@ int fec_dZ( const std::string & input_filename, std::fflush( stdout ); } unsigned long combinations = 0, repair_attempts = 0, successes = 0, failed_comparisons = 0; - int pct = ( prodata_size >= 1000 && isatty( STDERR_FILENO ) ) ? 0 : 100; - long pct_pos = ( pct < 100 ) ? 0 : prodata_size; + int pct = (prodata_size >= 1000 && isatty( STDERR_FILENO )) ? 0 : 100; + long pct_pos = (pct < 100) ? 0 : prodata_size; const int saved_verbosity = verbosity; verbosity = -1; // suppress all messages for( long pos = 0; pos <= last_pos; pos += delta ) @@ -108,7 +108,7 @@ bool check_inverse( const uint16_t * const A, const uint16_t * const B, { if( print && row ) std::fputc( '\n', stderr ); return false; } if( print ) std::fprintf( stderr, "\r%5u rows checked \r", row + 1 ); } - return true; // A * B == I + return true; } @@ -57,7 +57,7 @@ struct Galois8_table // addition/subtraction is exclusive or { uint8_t * const mul_row = mul_table + i * size; for( int j = 1; j < size; ++j ) - mul_row[j] = ilog[(log[i] + log[j]) % (size - 1)]; + mul_row[j] = ilog[(log[i] + log[j]) % (size-1)]; } for( int i = 0; i < size; ++i ) mul_table[0 * size + i] = mul_table[i * size + 0] = 0; @@ -79,7 +79,7 @@ bool check_inverse( const uint8_t * const A, const uint8_t * const B, uint8_t sum = 0; for( unsigned i = 0; i < k; ++i, ++pa, pb += k ) sum ^= gf.mul_table[*pa * gf.size + *pb]; - if( sum != ( row == col ) ) return false; // A * B != I + if( sum != ( row == col ) ) return false; } return true; } @@ -17,6 +17,7 @@ #define _FILE_OFFSET_BITS 64 +#include <cerrno> #include <cstdio> #include <cstring> #include <string> @@ -57,7 +58,7 @@ int list_files( const std::vector< std::string > & filenames, for( unsigned i = 0; i < filenames.size(); ++i ) { - const bool from_stdin = ( filenames[i] == "-" ); + const bool from_stdin = filenames[i] == "-"; if( from_stdin ) { if( stdin_used ) continue; else stdin_used = true; } const char * const input_filename = from_stdin ? "(stdin)" : filenames[i].c_str(); @@ -75,6 +76,9 @@ int list_files( const std::vector< std::string > & filenames, set_retval( retval, lzip_index.retval() ); continue; } + const bool multi_empty = + !from_stdin && !cl_opts.ignore_errors && lzip_index.multi_empty(); + if( multi_empty ) set_retval( retval, 2 ); if( verbosity < 0 ) continue; const unsigned long long udata_size = lzip_index.udata_size(); const unsigned long long cdata_size = lzip_index.cdata_size(); @@ -86,6 +90,8 @@ int list_files( const std::vector< std::string > & filenames, if( verbosity >= 1 ) std::fputs( " dict memb trail ", stdout ); std::fputs( " uncompressed compressed saved name\n", stdout ); } + if( multi_empty ) + { std::fflush( stdout ); show_file_error( input_filename, empty_msg ); } if( verbosity >= 1 ) std::printf( "%s %5ld %6lld ", format_ds( lzip_index.dictionary_size() ), members, lzip_index.file_size() - cdata_size ); @@ -113,12 +119,16 @@ int list_files( const std::vector< std::string > & filenames, first_post = true; // reprint heading after list of members } std::fflush( stdout ); + if( std::ferror( stdout ) ) break; } - if( verbosity >= 0 && files > 1 ) + if( verbosity >= 0 && files > 1 && !std::ferror( stdout ) ) { if( verbosity >= 1 ) std::fputs( " ", stdout ); list_line( total_uncomp, total_comp, "(totals)" ); std::fflush( stdout ); } + if( verbosity >= 0 && ( std::ferror( stdout ) || std::fclose( stdout ) != 0 ) ) + { show_file_error( "(stdout)", write_error_msg, errno ); + set_retval( retval, 1 ); } return retval; } diff --git a/lunzcrash.cc b/lunzcrash.cc index a1360cf..25d7c39 100644 --- a/lunzcrash.cc +++ b/lunzcrash.cc @@ -56,7 +56,7 @@ bool compare_member( const uint8_t * const mbuffer, const long msize, { MD5SUM md5sum; LZ_mtester mtester( mbuffer, msize, dictionary_size, -1, &md5sum ); - bool error = ( mtester.test_member() != 0 || !mtester.finished() ); + bool error = mtester.test_member() != 0 || !mtester.finished(); if( !error ) { md5_type new_digest; @@ -103,22 +103,23 @@ long next_pct_pos( const Lzip_index & lzip_index, const long i, const int pct, /* Test 1-bit errors in LZMA streams in file. Unless verbosity >= 1, print only the bytes with interesting results. */ -int lunzcrash_bit( const char * const input_filename, +int lunzcrash_bit( const std::string & input_filename, const Cl_options & cl_opts ) { + const char * const filename = input_filename.c_str(); struct stat in_stats; // not used - const int infd = open_instream( input_filename, &in_stats, false, true ); + const int infd = open_instream( filename, &in_stats, false, true ); if( infd < 0 ) return 1; const Lzip_index lzip_index( infd, cl_opts ); if( lzip_index.retval() != 0 ) - { show_file_error( input_filename, lzip_index.error().c_str() ); + { show_file_error( filename, lzip_index.error().c_str() ); return lzip_index.retval(); } - if( verbosity >= 2 ) printf( "Testing file '%s'\n", input_filename ); + if( verbosity >= 2 ) printf( "Testing file '%s'\n", filename ); const long long cdata_size = lzip_index.cdata_size(); long positions = 0, decompressions = 0, successes = 0, failed_comparisons = 0; - int pct = ( cdata_size >= 1000 && isatty( STDERR_FILENO ) ) ? 0 : 100; + int pct = (cdata_size >= 1000 && isatty( STDERR_FILENO )) ? 0 : 100; for( long i = 0; i < lzip_index.members(); ++i ) { const long long mpos = lzip_index.mblock( i ).pos(); @@ -127,8 +128,8 @@ int lunzcrash_bit( const char * const input_filename, if( !mbuffer ) return 1; const unsigned dictionary_size = lzip_index.dictionary_size( i ); md5_type md5_orig; - if( !check_member( mbuffer, msize, dictionary_size, input_filename, - md5_orig ) ) return 2; + if( !check_member( mbuffer, msize, dictionary_size, filename, md5_orig ) ) + return 2; long pct_pos = next_pct_pos( lzip_index, i, pct ); long pos = Lzip_header::size + 1, printed = 0; // last pos printed const long end = msize - 20; @@ -224,22 +225,23 @@ int lunzcrash_bit( const char * const input_filename, /* Test zeroed blocks of given size in LZMA streams in file. Unless verbosity >= 1, print only the bytes with interesting results. */ -int lunzcrash_block( const char * const input_filename, +int lunzcrash_block( const std::string & input_filename, const Cl_options & cl_opts, const int sector_size ) { + const char * const filename = input_filename.c_str(); struct stat in_stats; // not used - const int infd = open_instream( input_filename, &in_stats, false, true ); + const int infd = open_instream( filename, &in_stats, false, true ); if( infd < 0 ) return 1; const Lzip_index lzip_index( infd, cl_opts ); if( lzip_index.retval() != 0 ) - { show_file_error( input_filename, lzip_index.error().c_str() ); + { show_file_error( filename, lzip_index.error().c_str() ); return lzip_index.retval(); } - if( verbosity >= 2 ) printf( "Testing file '%s'\n", input_filename ); + if( verbosity >= 2 ) printf( "Testing file '%s'\n", filename ); const long long cdata_size = lzip_index.cdata_size(); long decompressions = 0, successes = 0, failed_comparisons = 0; - int pct = ( cdata_size >= 1000 && isatty( STDERR_FILENO ) ) ? 0 : 100; + int pct = (cdata_size >= 1000 && isatty( STDERR_FILENO )) ? 0 : 100; uint8_t * const block = new uint8_t[sector_size]; for( long i = 0; i < lzip_index.members(); ++i ) { @@ -251,8 +253,8 @@ int lunzcrash_block( const char * const input_filename, if( !mbuffer ) return 1; const unsigned dictionary_size = lzip_index.dictionary_size( i ); md5_type md5_orig; - if( !check_member( mbuffer, msize, dictionary_size, input_filename, - md5_orig ) ) return 2; + if( !check_member( mbuffer, msize, dictionary_size, filename, md5_orig ) ) + return 2; long pct_pos = next_pct_pos( lzip_index, i, pct, sector_size ); long pos = Lzip_header::size + 1; const long end = msize - sector_size - 20; @@ -344,7 +346,7 @@ int md5sum_files( const std::vector< std::string > & filenames ) for( unsigned i = 0; i < filenames.size(); ++i ) { - const bool from_stdin = ( filenames[i] == "-" ); + const bool from_stdin = filenames[i] == "-"; if( from_stdin ) { if( stdin_used ) continue; else stdin_used = true; } const char * const input_filename = filenames[i].c_str(); struct stat in_stats; // not used @@ -359,7 +361,8 @@ int md5sum_files( const std::vector< std::string > & filenames ) while( true ) { const int len = readblock( infd, buffer, buffer_size ); - if( len != buffer_size && errno ) throw Error( read_error_msg ); + if( len != buffer_size && errno ) + { show_file_error( input_filename, read_error_msg, errno ); return 1; } if( len > 0 ) md5sum.md5_update( buffer, len ); if( len < buffer_size ) break; } @@ -367,10 +370,15 @@ int md5sum_files( const std::vector< std::string > & filenames ) if( close( infd ) != 0 ) { show_file_error( input_filename, "Error closing input file", errno ); return 1; } + if( verbosity < 0 ) continue; for( int i = 0; i < 16; ++i ) std::printf( "%02x", md5_digest[i] ); std::printf( " %s\n", input_filename ); std::fflush( stdout ); + if( std::ferror( stdout ) ) break; } + if( verbosity >= 0 && ( std::ferror( stdout ) || std::fclose( stdout ) != 0 ) ) + { show_file_error( "(stdout)", write_error_msg, errno ); + set_retval( retval, 1 ); } return retval; } @@ -321,15 +321,12 @@ struct Lzip_trailer struct Cl_options // command-line options { - bool ignore_empty; bool ignore_errors; - bool ignore_nonzero; bool ignore_trailing; bool loose_trailing; - Cl_options() - : ignore_empty( false ), ignore_errors( false ), ignore_nonzero( false ), - ignore_trailing( true ), loose_trailing( false ) {} + Cl_options() : ignore_errors( false ), + ignore_trailing( true ), loose_trailing( false ) {} }; @@ -420,34 +417,44 @@ inline unsigned long long positive_diff( const unsigned long long x, inline void set_retval( int & retval, const int new_val ) { if( retval < new_val ) retval = new_val; } +inline const char * printable_name( const std::string & filename, + const bool in = true ) + { + if( filename.empty() || filename == "-" ) return in ? "(stdin)" : "(stdout)"; + return filename.c_str(); + } + const char * const bad_magic_msg = "Bad magic number (file not in lzip format)."; const char * const bad_dict_msg = "Invalid dictionary size in member header."; const char * const corrupt_mm_msg = "Corrupt header in multimember file."; const char * const empty_msg = "Empty member not allowed."; -const char * const nonzero_msg = "Nonzero first LZMA byte."; -const char * const trailing_msg = "Trailing data not allowed."; const char * const mmap_msg = "Can't mmap"; +const char * const nonzero_msg = "Nonzero first LZMA byte."; const char * const short_file_msg = "Input file is too short."; +const char * const trailing_msg = "Trailing data not allowed."; +const char * const write_error_msg = "Write error"; // defined in alone_to_lz.cc int alone_to_lz( const int infd, const Pretty_print & pp ); // defined in byte_repair.cc +bool safe_seek( const int fd, const long long pos, + const std::string & filename ); long seek_write( const int fd, const uint8_t * const buf, const long size, const long long pos ); uint8_t * read_member( const int infd, const long long mpos, - const long long msize, const char * const filename ); + const long long msize, const std::string & filename ); int byte_repair( const std::string & input_filename, const std::string & default_output_filename, const Cl_options & cl_opts, const char terminator, const bool force ); -int debug_delay( const char * const input_filename, +int debug_delay( const std::string & input_filename, const Cl_options & cl_opts, Block range, const char terminator ); -int debug_byte_repair( const char * const input_filename, +int debug_byte_repair( const std::string & input_filename, const Cl_options & cl_opts, const Bad_byte & bad_byte, const char terminator ); -int debug_decompress( const char * const input_filename, +int debug_decompress( const std::string & input_filename, const Cl_options & cl_opts, const Bad_byte & bad_byte, const bool show_packets ); @@ -470,9 +477,9 @@ int list_files( const std::vector< std::string > & filenames, const Cl_options & cl_opts ); // defined in lunzcrash.cc -int lunzcrash_bit( const char * const input_filename, +int lunzcrash_bit( const std::string & input_filename, const Cl_options & cl_opts ); -int lunzcrash_block( const char * const input_filename, +int lunzcrash_block( const std::string & input_filename, const Cl_options & cl_opts, const int sector_size ); int md5sum_files( const std::vector< std::string > & filenames ); @@ -509,10 +516,11 @@ void show_dprogress( const unsigned long long cfile_size = 0, const Pretty_print * const p = 0 ); // defined in merge.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 std::string & iname, + const std::string & oname, const long long max_size = -1 ); int test_member_from_file( const int infd, const unsigned long long msize, - long long * const failure_posp = 0 ); + long long * const failure_posp = 0, + bool * const nonzerop = 0 ); int merge_files( const std::vector< std::string > & filenames, const std::string & default_output_filename, const Cl_options & cl_opts, const char terminator, @@ -526,8 +534,6 @@ int print_nrep_stats( const std::vector< std::string > & filenames, const char * format_num( unsigned long long num, unsigned long long limit = -1ULL, const int set_prefix = 0 ); -bool safe_seek( const int fd, const long long pos, - const char * const filename ); int range_decompress( const std::string & input_filename, const std::string & default_output_filename, const Cl_options & cl_opts, Block range, @@ -540,7 +546,7 @@ int reproduce_file( const std::string & input_filename, const char * const reference_filename, const Cl_options & cl_opts, const int lzip_level, const char terminator, const bool force ); -int debug_reproduce_file( const char * const input_filename, +int debug_reproduce_file( const std::string & input_filename, const char * const lzip_name, const char * const reference_filename, const Cl_options & cl_opts, const Block & range, diff --git a/lzip_index.cc b/lzip_index.cc index 332fe63..02e3d28 100644 --- a/lzip_index.cc +++ b/lzip_index.cc @@ -148,13 +148,12 @@ bool Lzip_index::skip_gap( const int fd, unsigned long long & pos, if( !cl_opts.ignore_trailing ) { error_ = trailing_msg; retval_ = 2; return false; } } - const unsigned long long data_size = trailer.data_size(); pos = ipos + i - member_size; // good member const unsigned dictionary_size = header.dictionary_size(); if( dictionary_size_ < dictionary_size ) dictionary_size_ = dictionary_size; - member_vector.push_back( Member( 0, data_size, pos, member_size, - dictionary_size ) ); + member_vector.push_back( Member( 0, trailer.data_size(), pos, + member_size, dictionary_size ) ); return true; } if( ipos == 0 ) @@ -197,7 +196,7 @@ Lzip_index::Lzip_index( const int infd, const Cl_options & cl_opts, retval_ = 2; return; } // pos always points to a header or to ( EOF || max_pos ) - unsigned long long pos = ( max_pos > 0 ) ? max_pos : insize; + unsigned long long pos = (max_pos > 0) ? max_pos : insize; while( pos >= min_member_size ) { Lzip_trailer trailer; @@ -221,13 +220,12 @@ Lzip_index::Lzip_index( const int infd, const Cl_options & cl_opts, continue; else return; } set_num_error( "Bad header at pos ", pos - member_size ); break; } - const unsigned long long data_size = trailer.data_size(); pos -= member_size; // good member const unsigned dictionary_size = header.dictionary_size(); if( dictionary_size_ < dictionary_size ) dictionary_size_ = dictionary_size; - member_vector.push_back( Member( 0, data_size, pos, member_size, - dictionary_size ) ); + member_vector.push_back( Member( 0, trailer.data_size(), pos, + member_size, dictionary_size ) ); } // block at pos == 0 must be a member unless shorter than min_member_size if( pos >= min_member_size || ( pos != 0 && !ignore_gaps ) || @@ -237,10 +235,6 @@ Lzip_index::Lzip_index( const int infd, const Cl_options & cl_opts, if( retval_ == 0 ) { error_ = "Can't create file index."; retval_ = 2; } return; } - if( !cl_opts.ignore_empty && member_vector.size() > 1 ) - for( unsigned long i = 0; i < member_vector.size(); ++i ) - if( member_vector[i].dblock.size() == 0 ) - { member_vector.clear(); error_ = empty_msg; retval_ = 2; return; } std::reverse( member_vector.begin(), member_vector.end() ); for( unsigned long i = 0; ; ++i ) { diff --git a/lzip_index.h b/lzip_index.h index 9c8bd2b..6cd9233 100644 --- a/lzip_index.h +++ b/lzip_index.h @@ -64,6 +64,14 @@ public: int retval() const { return retval_; } unsigned dictionary_size() const { return dictionary_size_; } + 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; + } + bool operator==( const Lzip_index & li ) const { if( retval_ || li.retval_ || insize != li.insize || @@ -36,6 +36,7 @@ #include <string> #include <vector> #include <fcntl.h> +#include <pthread.h> // pthread_t #include <stdint.h> // SIZE_MAX #include <unistd.h> #include <utime.h> @@ -104,8 +105,8 @@ enum Mode { m_none, m_alone_to_lz, m_byte_repair, m_check, m_debug_byte_repair, m_nrep_stats, m_range_dec, m_remove, m_reproduce, m_show_packets, m_split, m_strip, m_test, m_unzcrash_bit, m_unzcrash_block }; -/* Variable used in signal handler context. - It is not declared volatile because the handler never returns. */ +/* Variables used in signal handler context. + They are not declared volatile because the handler never returns. */ bool delete_output_on_interrupt = false; @@ -156,8 +157,6 @@ void show_help( const long num_online ) " --dump=<list>:d:e:t dump members, damaged/empty, tdata to stdout\n" " --remove=<list>:d:e:t remove members, tdata from files in place\n" " --strip=<list>:d:e:t copy files to stdout stripping members given\n" - " --ignore-empty ignore empty members in multimember files\n" - " --ignore-nonzero ignore a nonzero first LZMA byte\n" " --loose-trailing allow trailing data seeming corrupt header\n" " --nonzero-repair repair in place a nonzero first LZMA byte\n", num_online ); @@ -228,7 +227,7 @@ const char * format_ds( const unsigned dictionary_size ) const char * p = ""; const char * np = " "; unsigned num = dictionary_size; - bool exact = ( num % factor == 0 ); + bool exact = num % factor == 0; for( int i = 0; i < n && ( num > 9999 || ( exact && num >= factor ) ); ++i ) { num /= factor; if( num % factor != 0 ) exact = false; @@ -263,12 +262,12 @@ void Member_list::parse_ml( const char * const arg, if( len <= 7 && std::strncmp( "damaged", p, len ) == 0 ) { damaged = true; cl_opts.ignore_errors = true; goto next; } if( len <= 5 && std::strncmp( "empty", p, len ) == 0 ) - { empty = true; cl_opts.ignore_empty = true; goto next; } + { empty = true; goto next; } if( len <= 5 && std::strncmp( "tdata", p, len ) == 0 ) { tdata = true; cl_opts.ignore_trailing = true; goto next; } } { - const bool reverse = ( *p == 'r' ); + const bool reverse = *p == 'r'; if( reverse ) ++p; if( *p == '^' ) { ++p; if( reverse ) rin = false; else in = false; } std::vector< Block > * rvp = reverse ? &rrange_vector : &range_vector; @@ -322,7 +321,7 @@ const char * parse_range( const char * const arg, const char * const pn, range.pos( value ); if( tail[0] == 0 || tail[0] == ':' ) { range.size( INT64_MAX - value ); return tail; } - const bool is_size = ( tail[0] == ',' ); + const bool is_size = tail[0] == ','; if( sector_sizep && tail[1] == ',' ) { value = INT64_MAX - value; ++tail; } else value = getnum( tail + 1, pn, 0, 1, INT64_MAX, &tail ); // size if( !is_size && value <= range.pos() ) @@ -390,6 +389,12 @@ void parse_range_vector( const char * const arg, const char * const pn, } +void no_to_stdout( const bool to_stdout ) + { + if( to_stdout ) + { show_error( "'--stdout' not allowed." ); std::exit( 1 ); } + } + void one_file( const int files ) { if( files != 1 ) @@ -562,9 +567,9 @@ int open_instream( const char * const name, struct stat * const in_statsp, { const int i = fstat( infd, in_statsp ); const mode_t mode = in_statsp->st_mode; - const bool can_read = ( i == 0 && !reg_only && - ( S_ISBLK( mode ) || S_ISCHR( mode ) || - S_ISFIFO( mode ) || S_ISSOCK( mode ) ) ); + const bool can_read = i == 0 && !reg_only && + ( S_ISBLK( mode ) || S_ISCHR( mode ) || + S_ISFIFO( mode ) || S_ISSOCK( mode ) ); if( i != 0 || ( !S_ISREG( mode ) && ( !can_read || one_to_one ) ) ) { if( verbosity >= 0 ) @@ -677,8 +682,8 @@ void set_signals( void (*action)(int) ) void cleanup_and_fail( const int retval ) { - cleanup_mutex_lock(); // only one thread can delete and exit set_signals( SIG_IGN ); // ignore signals + cleanup_mutex_lock(); // only one thread can delete and exit if( delete_output_on_interrupt ) { delete_output_on_interrupt = false; @@ -795,12 +800,12 @@ bool show_trailing_data( const uint8_t * const data, const int size, int decompress( const unsigned long long cfile_size, const int infd, const Cl_options & cl_opts, const Pretty_print & pp, - const bool testing ) + const bool from_stdin, const bool testing ) { unsigned long long partial_file_pos = 0; Range_decoder rdec( infd ); int retval = 0; - bool empty = false, nonempty = false; + bool empty = false, multi = false; for( bool first_member = true; ; first_member = false ) { @@ -843,11 +848,10 @@ int decompress( const unsigned long long cfile_size, const int infd, LZ_decoder decoder( rdec, dictionary_size, outfd ); show_dprogress( cfile_size, partial_file_pos, &rdec, &pp ); // init - const int result = decoder.decode_member( pp, cl_opts.ignore_nonzero ); + const int result = decoder.decode_member( pp, cl_opts.ignore_errors ); partial_file_pos += rdec.member_position(); if( result != 0 ) { - retval = 2; if( verbosity >= 0 && result <= 2 ) { pp(); @@ -855,19 +859,20 @@ int decompress( const unsigned long long cfile_size, const int infd, "File ends unexpectedly" : "Decoder error", partial_file_pos ); } - else if( result == 5 ) { pp( nonzero_msg ); break; } + else if( result == 5 ) pp( nonzero_msg ); + retval = 2; if( cl_opts.ignore_errors ) { pp.reset(); continue; } else break; } - if( !cl_opts.ignore_empty ) - { if( decoder.data_position() == 0 ) empty = true; else nonempty = true; } + if( !from_stdin && !cl_opts.ignore_errors ) { multi = !first_member; + if( decoder.data_position() == 0 ) empty = true; } if( verbosity >= 2 ) { std::fputs( testing ? "ok\n" : "done\n", stderr ); pp.reset(); } } if( verbosity == 1 && retval == 0 ) std::fputs( testing ? "ok\n" : "done\n", stderr ); - if( retval == 2 && cl_opts.ignore_errors ) retval = 0; - if( empty && nonempty && retval == 0 ) + if( empty && multi && retval == 0 ) { show_file_error( pp.name(), empty_msg ); retval = 2; } + if( retval == 2 && cl_opts.ignore_errors ) retval = 0; return retval; } @@ -972,8 +977,8 @@ int main( const int argc, const char * const argv[] ) bool to_stdout = false; if( argc > 0 ) invocation_name = argv[0]; - enum { opt_chk = 256, opt_dbg, opt_du, opt_ff, opt_g16, opt_ie, opt_inz, - opt_lt, opt_lzl, opt_lzn, opt_nzr, opt_ref, opt_rem, opt_rnd, opt_st }; + enum { opt_chk = 256, opt_dbg, opt_du, opt_ff, opt_g16, opt_lt, + opt_lzl, opt_lzn, opt_nzr, opt_ref, opt_rem, opt_rnd, opt_st }; const Arg_parser::Option options[] = { { '0', 0, Arg_parser::no }, @@ -1024,8 +1029,6 @@ int main( const int argc, const char * const argv[] ) { opt_du, "dump", Arg_parser::yes }, { opt_ff, "fec-file", Arg_parser::yes }, { opt_g16, "gf16", Arg_parser::no }, - { opt_ie, "ignore-empty", Arg_parser::no }, - { opt_inz, "ignore-nonzero", Arg_parser::no }, { opt_lt, "loose-trailing", Arg_parser::no }, { opt_lzl, "lzip-level", Arg_parser::yes }, { opt_lzn, "lzip-name", Arg_parser::yes }, @@ -1106,8 +1109,6 @@ int main( const int argc, const char * const argv[] ) member_list.parse_ml( arg, pn, cl_opts ); break; case opt_ff: cl_fec_filename = sarg; break; case opt_g16: cl_gf16 = true; break; - case opt_ie: cl_opts.ignore_empty = true; break; - case opt_inz: cl_opts.ignore_nonzero = true; break; case opt_lt: cl_opts.loose_trailing = true; break; case opt_lzl: lzip_level = parse_lzip_level( arg, pn ); break; case opt_lzn: lzip_name = arg; break; @@ -1132,9 +1133,6 @@ int main( const int argc, const char * const argv[] ) show_error( "You must specify the operation to be performed.", 0, true ); return 1; } - if( program_mode != m_decompress && program_mode != m_list && - program_mode != m_test && program_mode != m_range_dec ) - cl_opts.ignore_empty = true; std::vector< std::string > filenames; bool filenames_given = false; @@ -1151,20 +1149,19 @@ int main( const int argc, const char * const argv[] ) case m_none: internal_error( "invalid operation." ); break; case m_alone_to_lz: break; case m_byte_repair: - one_file( filenames.size() ); + one_file( filenames.size() ); no_to_stdout( to_stdout ); return byte_repair( filenames[0], default_output_filename, cl_opts, terminator, force ); case m_check: return gf_check( cblocks, cl_gf16, fec_random ); case m_debug_byte_repair: one_file( filenames.size() ); - return debug_byte_repair( filenames[0].c_str(), cl_opts, bad_byte, - terminator ); + return debug_byte_repair( filenames[0], cl_opts, bad_byte, terminator ); case m_debug_decompress: one_file( filenames.size() ); - return debug_decompress( filenames[0].c_str(), cl_opts, bad_byte, false ); + return debug_decompress( filenames[0], cl_opts, bad_byte, false ); case m_debug_delay: one_file( filenames.size() ); - return debug_delay( filenames[0].c_str(), cl_opts, range, terminator ); + return debug_delay( filenames[0], cl_opts, range, terminator ); case m_decompress: break; case m_dump: case m_strip: @@ -1197,7 +1194,7 @@ int main( const int argc, const char * const argv[] ) return fec_dZ( filenames[0], cl_fec_filename, delta, sector_size ); case m_list: break; case m_md5sum: break; - case m_merge: + case m_merge: no_to_stdout( to_stdout ); if( filenames.size() < 2 ) { show_error( "You must specify at least 2 files.", 0, true ); return 1; } return merge_files( filenames, default_output_filename, cl_opts, @@ -1215,28 +1212,28 @@ int main( const int argc, const char * const argv[] ) at_least_one_file( filenames.size() ); return remove_members( filenames, cl_opts, member_list ); case m_reproduce: - one_file( filenames.size() ); + one_file( filenames.size() ); no_to_stdout( to_stdout ); if( !reference_filename || !reference_filename[0] ) { show_error( "You must specify a reference file.", 0, true ); return 1; } if( range.size() > 0 ) - return debug_reproduce_file( filenames[0].c_str(), lzip_name, + return debug_reproduce_file( filenames[0], lzip_name, reference_filename, cl_opts, range, sector_size, lzip_level ); else return reproduce_file( filenames[0], default_output_filename, lzip_name, reference_filename, cl_opts, lzip_level, terminator, force ); case m_show_packets: one_file( filenames.size() ); - return debug_decompress( filenames[0].c_str(), cl_opts, bad_byte, true ); + return debug_decompress( filenames[0], cl_opts, bad_byte, true ); case m_split: - one_file( filenames.size() ); + one_file( filenames.size() ); no_to_stdout( to_stdout ); return split_file( filenames[0], default_output_filename, cl_opts, force ); case m_test: break; case m_unzcrash_bit: one_file( filenames.size() ); - return lunzcrash_bit( filenames[0].c_str(), cl_opts ); + return lunzcrash_bit( filenames[0], cl_opts ); case m_unzcrash_block: one_file( filenames.size() ); - return lunzcrash_block( filenames[0].c_str(), cl_opts, sector_size ); + return lunzcrash_block( filenames[0], cl_opts, sector_size ); } } catch( std::bad_alloc & ) { show_error( mem_msg ); cleanup_and_fail( 1 ); } @@ -1274,9 +1271,10 @@ int main( const int argc, const char * const argv[] ) { std::string input_filename; int infd; + const bool from_stdin = filenames[i] == "-"; pp.set_name( filenames[i] ); - if( filenames[i] == "-" ) + if( from_stdin ) { if( stdin_used ) continue; else stdin_used = true; infd = STDIN_FILENO; @@ -1318,7 +1316,8 @@ int main( const int argc, const char * const argv[] ) if( program_mode == m_alone_to_lz ) tmp = alone_to_lz( infd, pp ); else - tmp = decompress( cfile_size, infd, cl_opts, pp, program_mode == m_test ); + tmp = decompress( cfile_size, infd, cl_opts, pp, from_stdin, + program_mode == m_test ); } catch( std::bad_alloc & ) { pp( mem_msg ); tmp = 1; } catch( Error & e ) { pp(); show_error( e.msg, errno ); tmp = 1; } @@ -140,25 +140,23 @@ bool diff_member( const long long mpos, const long long msize, continue; std::vector< Block > bv; long long partial_pos = 0; - const char * const filename1 = filenames[i1].c_str(); - const char * const filename2 = filenames[i2].c_str(); const int fd1 = infd_vector[i1], fd2 = infd_vector[i2]; int begin = -1; // begin of block. -1 means no block bool prev_equal = true; - if( !safe_seek( fd1, mpos, filename1 ) || - !safe_seek( fd2, mpos, filename2 ) ) { error = true; break; } + if( !safe_seek( fd1, mpos, filenames[i1] ) || + !safe_seek( fd2, mpos, filenames[i2] ) ) { error = true; break; } while( partial_pos < msize ) { const int size = std::min( (long long)buffer_size, msize - partial_pos ); const int rd = readblock( fd1, buffer1, size ); if( rd != size && errno ) - { show_file_error( filename1, read_error_msg, errno ); + { show_file_error( filenames[i1].c_str(), read_error_msg, errno ); error = true; break; } if( rd > 0 ) { if( readblock( fd2, buffer2, rd ) != rd ) - { show_file_error( filename2, read_error_msg, errno ); + { show_file_error( filenames[i2].c_str(), read_error_msg, errno ); error = true; break; } for( int i = 0; i < rd; ++i ) { @@ -285,21 +283,20 @@ int open_input_files( const std::vector< std::string > & filenames, for( int i = 0; i < files; ++i ) { - const char * const filename = filenames[i].c_str(); const int infd = infd_vector[i]; bool error = false; for( long j = 0; j < lzip_index.members(); ++j ) { const long long mpos = lzip_index.mblock( j ).pos(); const long long msize = lzip_index.mblock( j ).size(); - if( !safe_seek( infd, mpos, filename ) ) return 1; + if( !safe_seek( infd, mpos, filenames[i] ) ) return 1; if( test_member_from_file( infd, msize ) != 0 ) { error = true; break; } } if( !error ) { if( verbosity >= 1 ) std::printf( "Input file '%s' has no errors. Recovery is not needed.\n", - filename ); + filenames[i].c_str() ); return 0; } } @@ -365,10 +362,10 @@ bool try_merge_member2( const std::vector< std::string > & filenames, if( i1 == i2 || color_vector[i1] == color_vector[i2] || color_done( color_vector, i1 ) ) continue; for( int bi = 0; bi < blocks; ++bi ) - if( !safe_seek( infd_vector[i2], block_vector[bi].pos(), filenames[i2].c_str() ) || - !safe_seek( outfd, block_vector[bi].pos(), output_filename.c_str() ) || - !copy_file( infd_vector[i2], outfd, block_vector[bi].size() ) ) - cleanup_and_fail( 1 ); + if( !safe_seek( infd_vector[i2], block_vector[bi].pos(), filenames[i2] ) || + !safe_seek( outfd, block_vector[bi].pos(), output_filename ) || + !copy_file( infd_vector[i2], outfd, filenames[i2], output_filename, + block_vector[bi].size() ) ) cleanup_and_fail( 1 ); const int infd = infd_vector[i1]; const int var = ( i1 * ( files - 1 ) ) + i2 - ( i2 > i1 ) + 1; for( int bi = 0; bi + 1 < blocks; ++bi ) @@ -379,10 +376,11 @@ bool try_merge_member2( const std::vector< std::string > & filenames, var, variations, bi + 1, terminator ); std::fflush( stdout ); pending_newline = true; } - if( !safe_seek( infd, block_vector[bi].pos(), filenames[i1].c_str() ) || - !safe_seek( outfd, block_vector[bi].pos(), output_filename.c_str() ) || - !copy_file( infd, outfd, block_vector[bi].size() ) || - !safe_seek( outfd, mpos, output_filename.c_str() ) ) + if( !safe_seek( infd, block_vector[bi].pos(), filenames[i1] ) || + !safe_seek( outfd, block_vector[bi].pos(), output_filename ) || + !copy_file( infd, outfd, filenames[i1], output_filename, + block_vector[bi].size() ) || + !safe_seek( outfd, mpos, output_filename ) ) cleanup_and_fail( 1 ); long long failure_pos = 0; if( test_member_from_file( outfd, msize, &failure_pos ) == 0 ) @@ -421,8 +419,7 @@ bool try_merge_member( const std::vector< std::string > & filenames, if( verbosity >= 2 ) { long var = 0; - for( int i = 0; i < blocks; ++i ) - var = ( var * files ) + file_idx[i]; + for( int i = 0; i < blocks; ++i ) var = var * files + file_idx[i]; std::printf( " Trying variation %ld of %ld %c", var + 1, variations, terminator ); std::fflush( stdout ); pending_newline = true; @@ -430,14 +427,13 @@ bool try_merge_member( const std::vector< std::string > & filenames, while( bi < blocks ) { const int infd = infd_vector[file_idx[bi]]; - if( !safe_seek( infd, block_vector[bi].pos(), filenames[file_idx[bi]].c_str() ) || - !safe_seek( outfd, block_vector[bi].pos(), output_filename.c_str() ) || - !copy_file( infd, outfd, block_vector[bi].size() ) ) - cleanup_and_fail( 1 ); + if( !safe_seek( infd, block_vector[bi].pos(), filenames[file_idx[bi]] ) || + !safe_seek( outfd, block_vector[bi].pos(), output_filename ) || + !copy_file( infd, outfd, filenames[file_idx[bi]], output_filename, + block_vector[bi].size() ) ) cleanup_and_fail( 1 ); ++bi; } - if( !safe_seek( outfd, mpos, output_filename.c_str() ) ) - cleanup_and_fail( 1 ); + if( !safe_seek( outfd, mpos, output_filename ) ) cleanup_and_fail( 1 ); long long failure_pos = 0; if( test_member_from_file( outfd, msize, &failure_pos ) == 0 ) return true; while( bi > 0 && mpos + failure_pos < block_vector[bi-1].pos() ) --bi; @@ -473,12 +469,12 @@ bool try_merge_member1( const std::vector< std::string > & filenames, { if( i1 == i2 || color_vector[i1] == color_vector[i2] || color_done( color_vector, i1 ) ) continue; + if( !safe_seek( infd_vector[i1], pos, filenames[i1] ) || + !safe_seek( infd_vector[i2], pos, filenames[i2] ) || + !safe_seek( outfd, pos, output_filename ) || + !copy_file( infd_vector[i2], outfd, filenames[i2], output_filename, + size ) ) cleanup_and_fail( 1 ); const int infd = infd_vector[i1]; - if( !safe_seek( infd, pos, filenames[i1].c_str() ) || - !safe_seek( infd_vector[i2], pos, filenames[i2].c_str() ) || - !safe_seek( outfd, pos, output_filename.c_str() ) || - !copy_file( infd_vector[i2], outfd, size ) ) - cleanup_and_fail( 1 ); const int var = ( i1 * ( files - 1 ) ) + i2 - ( i2 > i1 ) + 1; for( long long i = 0; i + 1 < size; ++i ) { @@ -488,10 +484,10 @@ bool try_merge_member1( const std::vector< std::string > & filenames, var, variations, pos + i, terminator ); std::fflush( stdout ); pending_newline = true; } - if( !safe_seek( outfd, pos + i, output_filename.c_str() ) || + if( !safe_seek( outfd, pos + i, output_filename ) || readblock( infd, &byte, 1 ) != 1 || writeblock( outfd, &byte, 1 ) != 1 || - !safe_seek( outfd, mpos, output_filename.c_str() ) ) + !safe_seek( outfd, mpos, output_filename ) ) cleanup_and_fail( 1 ); long long failure_pos = 0; if( test_member_from_file( outfd, msize, &failure_pos ) == 0 ) @@ -508,11 +504,12 @@ bool try_merge_member1( const std::vector< std::string > & filenames, /* 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. 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 std::string & iname, + const std::string & oname, const long long max_size ) { const int buffer_size = 65536; // 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; uint8_t * const buffer = new uint8_t[buffer_size]; bool error = false; @@ -523,20 +520,22 @@ bool copy_file( const int infd, const int outfd, const long long max_size ) if( max_size >= 0 ) rest -= size; const int rd = readblock( infd, buffer, size ); if( rd != size && errno ) - { show_error( read_error_msg, errno ); error = true; break; } + { show_file_error( printable_name( iname ), read_error_msg, errno ); + error = true; break; } if( rd > 0 ) { const int wr = writeblock( outfd, buffer, rd ); if( wr != rd ) - { show_error( "Error writing output file", errno ); - error = true; break; } + { show_file_error( printable_name( oname, false ), write_error_msg, + errno ); error = true; break; } copied_size += rd; } if( rd < size ) break; // EOF } delete[] buffer; if( !error && max_size >= 0 && copied_size != max_size ) - { show_error( "Input file ends unexpectedly." ); error = true; } + { show_file_error( printable_name( iname ), "Input file ends unexpectedly." ); + error = true; } return !error; } @@ -544,7 +543,7 @@ bool copy_file( const int infd, const int outfd, const long long max_size ) /* Return value: 0 = OK, 1 = bad msize, 2 = data error. 'failure_pos' is relative to the beginning of the member. */ int test_member_from_file( const int infd, const unsigned long long msize, - long long * const failure_posp ) + long long * const failure_posp, bool * const nonzerop ) { Range_decoder rdec( infd ); Lzip_header header; @@ -559,6 +558,7 @@ int test_member_from_file( const int infd, const unsigned long long msize, verbosity = -1; // suppress all messages done = decoder.decode_member() == 0; verbosity = saved_verbosity; // restore verbosity level + if( nonzerop ) *nonzerop = rdec.nonzero(); if( done && rdec.member_position() == msize ) return 0; } if( failure_posp ) *failure_posp = rdec.member_position(); @@ -578,15 +578,15 @@ int merge_files( const std::vector< std::string > & filenames, const int retval = open_input_files( filenames, infd_vector, cl_opts, lzip_index, &in_stats ); if( retval >= 0 ) return retval; - if( !safe_seek( infd_vector[0], 0, filenames[0].c_str() ) ) return 1; + if( !safe_seek( infd_vector[0], 0, filenames[0] ) ) return 1; const bool to_file = default_output_filename.size(); output_filename = to_file ? default_output_filename : insert_fixed( filenames[0] ); set_signal_handler(); if( !open_outstream( force, true, true, false, to_file ) ) return 1; - if( !copy_file( infd_vector[0], outfd ) ) // copy whole file - cleanup_and_fail( 1 ); + if( !copy_file( infd_vector[0], outfd, filenames[0], output_filename ) ) + cleanup_and_fail( 1 ); // copy whole file for( long j = 0; j < lzip_index.members(); ++j ) { @@ -597,7 +597,7 @@ int merge_files( const std::vector< std::string > & filenames, // different color means members are different std::vector< int > color_vector( files, 0 ); if( !diff_member( mpos, msize, filenames, infd_vector, block_vector, - color_vector ) || !safe_seek( outfd, mpos, output_filename.c_str() ) ) + color_vector ) || !safe_seek( outfd, mpos, output_filename ) ) cleanup_and_fail( 1 ); if( block_vector.empty() ) @@ -83,7 +83,7 @@ void LZ_mtester::flush_data() crc32.update_buf( crc_, buffer + stream_pos, size ); if( md5sum ) md5sum->md5_update( buffer + stream_pos, size ); if( outfd >= 0 && writeblock( outfd, buffer + stream_pos, size ) != size ) - throw Error( "Write error" ); + throw Error( write_error_msg ); if( pos >= dictionary_size ) { partial_data_pos += pos; pos = 0; pos_wrapped = true; } stream_pos = pos; @@ -148,7 +148,7 @@ int LZ_mtester::test_member( const unsigned long mpos_limit, FILE * const f, const unsigned long long byte_pos ) { if( mpos_limit < Lzip_header::size + 5 ) return -1; - if( member_position() == Lzip_header::size ) rdec.load(); + if( member_position() == Lzip_header::size && !rdec.load() ) return 1; while( !rdec.finished() ) { if( member_position() >= mpos_limit || data_position() >= dpos_limit ) @@ -194,22 +194,22 @@ int LZ_mtester::test_member( const unsigned long mpos_limit, } else // match { + rep3 = rep2; rep2 = rep1; rep1 = rep0; len = rdec.decode_len( match_len_model, pos_state ); - unsigned distance = rdec.decode_tree6( bm_dis_slot[get_len_state(len)] ); - if( distance >= start_dis_model ) + rep0 = rdec.decode_tree6( bm_dis_slot[get_len_state(len)] ); + if( rep0 >= start_dis_model ) { - const unsigned dis_slot = distance; + const unsigned dis_slot = rep0; const int direct_bits = ( dis_slot >> 1 ) - 1; - distance = ( 2 | ( dis_slot & 1 ) ) << direct_bits; + rep0 = ( 2 | ( dis_slot & 1 ) ) << direct_bits; if( dis_slot < end_dis_model ) - distance += rdec.decode_tree_reversed( - bm_dis + ( distance - dis_slot ), direct_bits ); + rep0 += rdec.decode_tree_reversed( bm_dis + ( rep0 - dis_slot ), + direct_bits ); else { - distance += - rdec.decode( direct_bits - dis_align_bits ) << dis_align_bits; - distance += rdec.decode_tree_reversed4( bm_align ); - if( distance == 0xFFFFFFFFU ) // marker found + rep0 += rdec.decode( direct_bits - dis_align_bits ) << dis_align_bits; + rep0 += rdec.decode_tree_reversed4( bm_align ); + if( rep0 == 0xFFFFFFFFU ) // marker found { rdec.normalize(); flush_data(); @@ -224,7 +224,6 @@ int LZ_mtester::test_member( const unsigned long mpos_limit, } } } - rep3 = rep2; rep2 = rep1; rep1 = rep0; rep0 = distance; if( rep0 > max_rep0 ) max_rep0 = rep0; state.set_match(); if( rep0 >= dictionary_size || ( rep0 >= pos && !pos_wrapped ) ) @@ -242,7 +241,7 @@ int LZ_mtester::test_member( const unsigned long mpos_limit, int LZ_mtester::debug_decode_member( const long long dpos, const long long mpos, const bool show_packets ) { - rdec.load(); + if( !rdec.load() ) return 1; unsigned old_tmpos = member_position(); // truncated member position while( !rdec.finished() ) { @@ -317,22 +316,22 @@ int LZ_mtester::debug_decode_member( const long long dpos, const long long mpos, } else // match { + rep3 = rep2; rep2 = rep1; rep1 = rep0; len = rdec.decode_len( match_len_model, pos_state ); - unsigned distance = rdec.decode_tree6( bm_dis_slot[get_len_state(len)] ); - if( distance >= start_dis_model ) + rep0 = rdec.decode_tree6( bm_dis_slot[get_len_state(len)] ); + if( rep0 >= start_dis_model ) { - const unsigned dis_slot = distance; + const unsigned dis_slot = rep0; const int direct_bits = ( dis_slot >> 1 ) - 1; - distance = ( 2 | ( dis_slot & 1 ) ) << direct_bits; + rep0 = ( 2 | ( dis_slot & 1 ) ) << direct_bits; if( dis_slot < end_dis_model ) - distance += rdec.decode_tree_reversed( - bm_dis + ( distance - dis_slot ), direct_bits ); + rep0 += rdec.decode_tree_reversed( bm_dis + ( rep0 - dis_slot ), + direct_bits ); else { - distance += - rdec.decode( direct_bits - dis_align_bits ) << dis_align_bits; - distance += rdec.decode_tree_reversed4( bm_align ); - if( distance == 0xFFFFFFFFU ) // marker found + rep0 += rdec.decode( direct_bits - dis_align_bits ) << dis_align_bits; + rep0 += rdec.decode_tree_reversed4( bm_align ); + if( rep0 == 0xFFFFFFFFU ) // marker found { rdec.normalize(); flush_data(); @@ -353,7 +352,6 @@ int LZ_mtester::debug_decode_member( const long long dpos, const long long mpos, } } } - rep3 = rep2; rep2 = rep1; rep1 = rep0; rep0 = distance; if( rep0 > max_rep0 ) { max_rep0 = rep0; max_rep0_pos = mp; } state.set_match(); if( show_packets ) @@ -53,12 +53,15 @@ public: return p; } - void load() + bool load() { code = 0; range = 0xFFFFFFFFU; + // check first byte of the LZMA stream without reading it + if( buffer[pos] != 0 ) return false; get_byte(); // discard first byte of the LZMA stream for( int i = 0; i < 4; ++i ) code = ( code << 8 ) | get_byte(); + return true; } void normalize() @@ -76,7 +79,7 @@ public: range >>= 1; // symbol <<= 1; // if( code >= range ) { code -= range; symbol |= 1; } - const bool bit = ( code >= range ); + const bool bit = code >= range; symbol <<= 1; symbol += bit; code -= range & ( 0U - bit ); } @@ -299,14 +302,14 @@ class LZ_mtester bool fast, fast2; if( lpos > distance ) { - fast = ( len < dictionary_size - lpos ); - fast2 = ( fast && len <= lpos - i ); + fast = len < dictionary_size - lpos; + fast2 = fast && len <= lpos - i; } else { i += dictionary_size; - fast = ( len < dictionary_size - i ); // (i == pos) may happen - fast2 = ( fast && len <= i - lpos ); + fast = len < dictionary_size - i; // (i == pos) may happen + fast2 = fast && len <= i - lpos; } if( fast ) // no wrap { @@ -378,8 +381,8 @@ public: const uint8_t * get_buffers( const uint8_t ** const prev_bufferp, int * const sizep, int * const prev_sizep ) const - { *sizep = ( pos_wrapped && pos == 0 ) ? dictionary_size : pos; - *prev_sizep = ( pos_wrapped && pos > 0 ) ? dictionary_size - pos : 0; + { *sizep = (pos_wrapped && pos == 0) ? dictionary_size : pos; + *prev_sizep = (pos_wrapped && pos > 0) ? dictionary_size - pos : 0; *prev_bufferp = buffer + pos; return buffer; } void duplicate_buffer( uint8_t * const buffer2 ); diff --git a/nrep_stats.cc b/nrep_stats.cc index 746a3bc..fda861a 100644 --- a/nrep_stats.cc +++ b/nrep_stats.cc @@ -44,11 +44,11 @@ int print_nrep_stats( const std::vector< std::string > & filenames, unsigned long long lzma_size = 0; // total size of LZMA data unsigned long best_pos = 0; int best_name = -1, retval = 0; - const bool count_all = ( repeated_byte < 0 || repeated_byte >= 256 ); + const bool count_all = repeated_byte < 0 || repeated_byte >= 256; bool stdin_used = false; for( unsigned i = 0; i < filenames.size(); ++i ) { - const bool from_stdin = ( filenames[i] == "-" ); + const bool from_stdin = filenames[i] == "-"; if( from_stdin ) { if( stdin_used ) continue; else stdin_used = true; } const char * const input_filename = from_stdin ? "(stdin)" : filenames[i].c_str(); diff --git a/range_dec.cc b/range_dec.cc index 0bb3078..23e18d5 100644 --- a/range_dec.cc +++ b/range_dec.cc @@ -53,7 +53,7 @@ bool decompress_member( const int infd, const Cl_options & cl_opts, if( verbosity >= 2 ) pp(); LZ_decoder decoder( rdec, dictionary_size, outfd, outskip, outend ); - const int result = decoder.decode_member( pp, cl_opts.ignore_nonzero ); + const int result = decoder.decode_member( pp, cl_opts.ignore_errors ); if( result != 0 ) { if( verbosity >= 0 && result <= 2 ) @@ -94,7 +94,7 @@ const char * format_num( unsigned long long num, static int current = 0; static bool si = true; - if( set_prefix ) si = ( set_prefix > 0 ); + if( set_prefix ) si = set_prefix > 0; unsigned long long den = 1; const unsigned factor = si ? 1000 : 1024; char * const buf = buffer[current++]; current %= buffers; @@ -112,15 +112,6 @@ const char * format_num( unsigned long long num, } -bool safe_seek( const int fd, const long long pos, - const char * const filename ) - { - if( lseek( fd, pos, SEEK_SET ) == pos ) return true; - show_file_error( filename, "Seek error", errno ); - return false; - } - - int range_decompress( const std::string & input_filename, const std::string & default_output_filename, const Cl_options & cl_opts, Block range, @@ -171,7 +162,7 @@ int range_decompress( const std::string & input_filename, const long long outskip = std::max( 0LL, range.pos() - db.pos() ); const long long outend = std::min( db.size(), range.end() - db.pos() ); const long long mpos = lzip_index.mblock( i ).pos(); - if( !safe_seek( infd, mpos, filename ) ) cleanup_and_fail( 1 ); + if( !safe_seek( infd, mpos, input_filename ) ) cleanup_and_fail( 1 ); if( !decompress_member( infd, cl_opts, pp, mpos, outskip, outend ) ) { if( cl_opts.ignore_errors ) error = true; else cleanup_and_fail( 2 ); } pp.reset(); diff --git a/recursive.cc b/recursive.cc index aaf84d7..a6e9317 100644 --- a/recursive.cc +++ b/recursive.cc @@ -45,7 +45,7 @@ bool test_full_name( const std::string & full_name, const struct stat * stp, if( !S_ISDIR( st.st_mode ) ) return false; std::string prev_dir( full_name ); - bool loop = ( stp && st.st_ino == stp->st_ino && st.st_dev == stp->st_dev ); + bool loop = stp && st.st_ino == stp->st_ino && st.st_dev == stp->st_dev; if( !loop ) for( unsigned i = prev_dir.size(); i > 1; ) { diff --git a/reproduce.cc b/reproduce.cc index ff7c654..d316499 100644 --- a/reproduce.cc +++ b/reproduce.cc @@ -459,7 +459,7 @@ int reproduce_member( uint8_t * const mbuffer, const long msize, if( offset < 0 ) { delete master; return 2; } // no match /* Reference data from offset must be at least as large as zeroed sector minus member trailer if trailer is inside the zeroed sector. */ - const int t = ( begin + size >= msize ) ? 16 + Lzip_trailer::size : 0; + const int t = (begin + size >= msize) ? 16 + Lzip_trailer::size : 0; if( rsize - offset < size - t ) { show_file_error( reference_filename, "Not enough reference data after match." ); delete master; return 2; } @@ -482,7 +482,7 @@ int reproduce_member( uint8_t * const mbuffer, const long msize, std::fflush( stdout ); pending_newline = true; } const bool level0 = level == '0'; - const bool auto0 = ( level0 && lzip_level != '0' ); + const bool auto0 = level0 && lzip_level != '0'; int ret = try_reproduce( mbuffer, msize, dsize, good_dsize, begin, end, rbuf, rsize, offset, dictionary_size, level0 ? lzip0_argv : lzip_argv, md5sump, terminator, auto0 ); @@ -550,7 +550,7 @@ int reproduce_file( const std::string & input_filename, i + 1, lzip_index.members(), terminator ); std::fflush( stdout ); pending_newline = true; } - if( !safe_seek( infd, mpos, filename ) ) return 1; + if( !safe_seek( infd, mpos, input_filename ) ) return 1; long long failure_pos = 0; if( test_member_from_file( infd, msize, &failure_pos ) == 0 ) continue; // member is not damaged @@ -592,11 +592,11 @@ int reproduce_file( const std::string & input_filename, { if( outfd < 0 ) // first damaged member reproduced { - if( !safe_seek( infd, 0, filename ) ) return 1; + if( !safe_seek( infd, 0, input_filename ) ) return 1; set_signal_handler(); if( !open_outstream( true, true, false, true, to_file ) ) return 1; - if( !copy_file( infd, outfd ) ) // copy whole file - cleanup_and_fail( 1 ); + if( !copy_file( infd, outfd, input_filename, output_filename ) ) + cleanup_and_fail( 1 ); // copy whole file } if( seek_write( outfd, mbuffer + begin, size, mpos + begin ) != size ) { show_file_error( output_filename.c_str(), "Error writing file", errno ); @@ -636,24 +636,25 @@ int reproduce_file( const std::string & input_filename, /* Passes a 0 terminator to other functions to prevent intramember feedback. Exits only in case of fatal error. (reference file too large, etc). */ -int debug_reproduce_file( const char * const input_filename, +int debug_reproduce_file( const std::string & input_filename, const char * const lzip_name, const char * const reference_filename, const Cl_options & cl_opts, const Block & range, const int sector_size, const int lzip_level ) { + const char * const filename = input_filename.c_str(); struct stat in_stats; // not used - const int infd = open_instream( input_filename, &in_stats, false, true ); + const int infd = open_instream( filename, &in_stats, false, true ); if( infd < 0 ) return 1; const Lzip_index lzip_index( infd, cl_opts ); if( lzip_index.retval() != 0 ) - { show_file_error( input_filename, lzip_index.error().c_str() ); + { show_file_error( filename, lzip_index.error().c_str() ); return lzip_index.retval(); } const long long cdata_size = lzip_index.cdata_size(); if( range.pos() >= cdata_size ) - { show_file_error( input_filename, "Range is beyond end of last member." ); + { show_file_error( filename, "Range is beyond end of last member." ); return 1; } const long page_size = std::max( 1L, sysconf( _SC_PAGESIZE ) ); @@ -670,7 +671,7 @@ int debug_reproduce_file( const char * const input_filename, const long long msize = lzip_index.mblock( i ).size(); if( !range.overlaps( mpos, msize ) ) continue; if( !fits_in_size_t( msize + page_size ) ) // mmap uses size_t - { show_file_error( input_filename, + { show_file_error( filename, "Input file contains member too large for mmap." ); return 1; } const long long dsize = lzip_index.dblock( i ).size(); const unsigned dictionary_size = lzip_index.dictionary_size( i ); @@ -687,14 +688,14 @@ int debug_reproduce_file( const char * const input_filename, uint8_t * const mbuffer_base = (uint8_t *)mmap( 0, msize + mpos_rem, PROT_READ | PROT_WRITE, MAP_PRIVATE, infd, mpos - mpos_rem ); if( mbuffer_base == MAP_FAILED ) - { show_file_error( input_filename, mmap_msg, errno ); return 1; } + { show_file_error( filename, mmap_msg, errno ); return 1; } uint8_t * const mbuffer = mbuffer_base + mpos_rem; if( !md5_valid ) { if( verbosity >= 0 ) // give a clue of the range being tested { std::printf( "Reproducing: %s\nReference file: %s\nTesting " "sectors of size %llu at file positions %llu to %llu\n", - input_filename, reference_filename, + filename, reference_filename, std::min( (long long)sector_size, rm_end - sector_pos ), sector_pos, rm_end - 1 ); std::fflush( stdout ); } md5_valid = true; compute_md5( mbuffer, msize, md5_digest_c ); @@ -719,7 +720,7 @@ int debug_reproduce_file( const char * const input_filename, long size = 0; uint8_t value = 0; const long begin = - zeroed_sector_pos( mbuffer, msize, input_filename, &size, &value ); + zeroed_sector_pos( mbuffer, msize, filename, &size, &value ); if( begin < 0 ) return 2; MD5SUM md5sum; const int ret = reproduce_member( mbuffer, msize, dsize, lzip_name, @@ -88,14 +88,14 @@ int split_file( const std::string & input_filename, long long mpos = b.pos(); long long msize = b.size(); long long failure_pos = 0; - if( !safe_seek( infd, mpos, filename ) ) return 1; + if( !safe_seek( infd, mpos, input_filename ) ) return 1; if( test_member_from_file( infd, msize, &failure_pos ) == 1 ) { // corrupt or fake trailer while( true ) { mpos += failure_pos; msize -= failure_pos; if( msize < min_member_size ) break; // trailing data - if( !safe_seek( infd, mpos, filename ) ) return 1; + if( !safe_seek( infd, mpos, input_filename ) ) return 1; if( test_member_from_file( infd, msize, &failure_pos ) != 1 ) break; } lzip_index = Lzip_index( infd, cl_opts, true, true, mpos ); @@ -106,7 +106,7 @@ int split_file( const std::string & input_filename, } } - if( !safe_seek( infd, 0, filename ) ) return 1; + if( !safe_seek( infd, 0, input_filename ) ) return 1; int max_digits = 1; for( long i = lzip_index.blocks( true ); i >= 10; i /= 10 ) ++max_digits; bool to_file = // if true, create intermediate dirs @@ -120,12 +120,13 @@ int split_file( const std::string & input_filename, if( mb.pos() > stream_pos ) // gap { if( !open_outstream( force, true, false, false, to_file ) ) return 1; - if( !copy_file( infd, outfd, mb.pos() - stream_pos ) || + if( !copy_file( infd, outfd, input_filename, output_filename, + mb.pos() - stream_pos ) || !close_outstream( &in_stats ) ) cleanup_and_fail( 1 ); next_filename( max_digits ); to_file = false; } if( !open_outstream( force, true, false, false, to_file ) ) return 1; // member - if( !copy_file( infd, outfd, mb.size() ) || + if( !copy_file( infd, outfd, input_filename, output_filename, mb.size() ) || !close_outstream( &in_stats ) ) cleanup_and_fail( 1 ); next_filename( max_digits ); to_file = false; stream_pos = mb.end(); @@ -133,7 +134,8 @@ int split_file( const std::string & input_filename, if( lzip_index.file_size() > stream_pos ) // trailing data { if( !open_outstream( force, true, false, false, to_file ) ) return 1; - if( !copy_file( infd, outfd, lzip_index.file_size() - stream_pos ) || + if( !copy_file( infd, outfd, input_filename, output_filename, + lzip_index.file_size() - stream_pos ) || !close_outstream( &in_stats ) ) cleanup_and_fail( 1 ); next_filename( max_digits ); to_file = false; } diff --git a/testsuite/check.sh b/testsuite/check.sh index 638d6db..30b2edb 100755 --- a/testsuite/check.sh +++ b/testsuite/check.sh @@ -32,7 +32,6 @@ cd "${objdir}"/tmp || framework_failure cp "${testdir}"/test.txt in || framework_failure in_lz="${testdir}"/test.txt.lz in_lzma="${testdir}"/test.txt.lzma -in_em="${testdir}"/test_em.txt.lz inD="${testdir}"/test21636.txt bad1_lz="${testdir}"/test_bad1.lz bad2_lz="${testdir}"/test_bad2.lz @@ -44,8 +43,9 @@ bad7_lz="${testdir}"/test_bad7.lz bad8_lz="${testdir}"/test_bad8.lz bad9_lz="${testdir}"/test_bad9.lz fox_lz="${testdir}"/fox.lz -fox6_lz="${testdir}"/fox6.lz +fnz_lz="${testdir}"/fox_nz.lz f6b1="${testdir}"/fox6_bad1.txt +f6b1nz_lz="${testdir}"/fox6_b1nz.lz f6b1_lz="${testdir}"/fox6_bad1.lz f6b2_lz="${testdir}"/fox6_bad2.lz f6b3_lz="${testdir}"/fox6_bad3.lz @@ -58,7 +58,6 @@ f6s3_lz="${testdir}"/fox6_sc3.lz f6s4_lz="${testdir}"/fox6_sc4.lz f6s5_lz="${testdir}"/fox6_sc5.lz f6s6_lz="${testdir}"/fox6_sc6.lz -f6nz_lz="${testdir}"/fox6_nz.lz num_lz="${testdir}"/numbers.lz nbt_lz="${testdir}"/numbersbt.lz fail=0 @@ -76,10 +75,8 @@ test_failed() { fail=1 ; printf " $1" ; [ -z "$2" ] || printf "($2)" ; } # test_bad8.lz: [ 66- 73] --> zeroed (reference test.txt [ 89- 110]) # test_bad9.lz: [6491-6498] --> zeroed (reference test.txt [17982-18594]) # -# test_em.txt.lz: test.txt split in 3, with 5 empty members (1,3,5-6,8) -# test_3m.txt.lz.md5: md5sum of test_em.txt.lz after removing empty members -# # 6-member files; fox6.lz with one or more errors +# fox6_b1nz.lz: fox6_bad1.lz with first LZMA byte set to 'a'..'f' # fox6_bad1.lz: byte at offset 5 changed from 0x0C to 0x00 (DS) # byte at offset 142 changed from 0x50 to 0x70 (CRC) # byte at offset 224 changed from 0x2D to 0x2E (data_size) @@ -92,8 +89,6 @@ test_failed() { fail=1 ; printf " $1" ; [ -z "$2" ] || printf "($2)" ; } # fox6_bad5.lz: [380-479] --> zeroed (members 5,6) # fox6_bad6.lz: [430-439] --> zeroed (member 6) # -# fox6_nz.lz: first LZMA byte of 4 last members set to 'a', 'b', 'c', 'd' -# # 6-member files "shortcircuited" by a corrupt or fake trailer # fox6_sc1.lz: (corrupt but consistent last trailer) # last CRC != 0 ; dsize = 4 * msize ; msize = 480 (file size) @@ -114,7 +109,8 @@ printf "testing lziprecover-%s..." "$2" "${LZIPRECOVER}" -q --nrep-stats=0N "${in_lz}" [ $? = 1 ] || test_failed $LINENO for i in 0 255 0kB 0KiB 0M 0G 0T 0P 0E 0Z 0Y 0R 0Q ; do - "${LZIPRECOVER}" -q --nrep-stats=$i "${in_lz}" || test_failed $LINENO $i + "${LZIPRECOVER}" --nrep-stats=$i "${in_lz}" > /dev/null || + test_failed $LINENO $i done "${LZIP}" -lq in [ $? = 2 ] || test_failed $LINENO @@ -249,6 +245,7 @@ cp "${in_lzma}" out.lzma || framework_failure cmp "${in_lz}" out.lz || test_failed $LINENO printf "to be overwritten" > out.lz || framework_failure "${LZIPRECOVER}" -Af out.lzma || test_failed $LINENO +[ ! -e out.lzma ] || test_failed $LINENO cmp "${in_lz}" out.lz || test_failed $LINENO rm -f out.lz || framework_failure cp "${in_lzma}" out.tlz || framework_failure @@ -261,39 +258,26 @@ rm -f out.tar.lz || framework_failure cat in in > in2 || framework_failure "${LZIPRECOVER}" -A -o out2.lz - "${in_lzma}" - < "${in_lzma}" || test_failed $LINENO -"${LZIP}" -cd out2.lz > out2 || test_failed $LINENO +"${LZIP}" -d out2.lz || test_failed $LINENO cmp in2 out2 || test_failed $LINENO -rm -f out2.lz out2 || framework_failure +rm -f out2 || framework_failure "${LZIPRECOVER}" -A "${in_lzma}" -o a/b/c/out.lz || test_failed $LINENO cmp "${in_lz}" a/b/c/out.lz || test_failed $LINENO rm -rf a || framework_failure printf "\ntesting decompression..." -for i in "${in_lz}" "${in_em}" ; do - "${LZIP}" -lq "$i" --ignore-empty || test_failed $LINENO "$i" - "${LZIP}" -t "$i" --ignore-empty || test_failed $LINENO "$i" - "${LZIP}" -d "$i" --ignore-empty -o out || test_failed $LINENO "$i" - cmp in out || test_failed $LINENO "$i" - "${LZIP}" -cd "$i" --ignore-empty > out || test_failed $LINENO "$i" - cmp in out || test_failed $LINENO "$i" - "${LZIP}" -d "$i" --ignore-empty -o - > out || test_failed $LINENO "$i" - cmp in out || test_failed $LINENO "$i" - "${LZIP}" -d < "$i" --ignore-empty > out || test_failed $LINENO "$i" - cmp in out || test_failed $LINENO "$i" - rm -f out || framework_failure -done - -lines=`"${LZIP}" -tvv "${in_em}" --ignore-empty 2>&1 | wc -l` || - test_failed $LINENO -[ "${lines}" -eq 8 ] || test_failed $LINENO "${lines}" -"${LZIP}" -tq "${in_em}" -[ $? = 2 ] || test_failed $LINENO - -lines=`"${LZIP}" -lvv "${in_em}" --ignore-empty | wc -l` || test_failed $LINENO -[ "${lines}" -eq 11 ] || test_failed $LINENO "${lines}" -"${LZIP}" -lq "${in_em}" -[ $? = 2 ] || test_failed $LINENO +"${LZIP}" -l "${in_lz}" > /dev/null || test_failed $LINENO +"${LZIP}" -t "${in_lz}" || test_failed $LINENO +"${LZIP}" -d "${in_lz}" -o out || test_failed $LINENO +cmp in out || test_failed $LINENO +"${LZIP}" -cd "${in_lz}" > out || test_failed $LINENO +cmp in out || test_failed $LINENO +"${LZIP}" -d "${in_lz}" -o - > out || test_failed $LINENO +cmp in out || test_failed $LINENO +"${LZIP}" -d < "${in_lz}" > out || test_failed $LINENO +cmp in out || test_failed $LINENO +rm -f out || framework_failure cp "${in_lz}" out.lz || framework_failure "${LZIP}" -dk out.lz || test_failed $LINENO @@ -315,7 +299,6 @@ rm -f copy out || framework_failure printf "to be overwritten" > out || framework_failure "${LZIP}" -df -o out < "${in_lz}" || test_failed $LINENO cmp in out || test_failed $LINENO -rm -f out || framework_failure "${LZIP}" -d -o ./- "${in_lz}" || test_failed $LINENO cmp in ./- || test_failed $LINENO rm -f ./- || framework_failure @@ -328,7 +311,7 @@ cp "${in_lz}" anyothername || framework_failure test_failed $LINENO cmp in out || test_failed $LINENO cmp in anyothername.out || test_failed $LINENO -rm -f out anyothername.out || framework_failure +rm -f anyothername.out || framework_failure "${LZIP}" -lq in "${in_lz}" [ $? = 2 ] || test_failed $LINENO @@ -365,7 +348,7 @@ done cmp in out || test_failed $LINENO rm -f out || framework_failure -"${LZIP}" -lq "${in_lz}" "${in_lz}" || test_failed $LINENO +"${LZIP}" -l "${in_lz}" "${in_lz}" > /dev/null || test_failed $LINENO "${LZIP}" -t "${in_lz}" "${in_lz}" || test_failed $LINENO "${LZIP}" -cd "${in_lz}" "${in_lz}" -o out > out2 || test_failed $LINENO [ ! -e out ] || test_failed $LINENO # override -o @@ -376,10 +359,13 @@ cmp in2 out2 || test_failed $LINENO rm -f out2 || framework_failure cat "${in_lz}" "${in_lz}" > out2.lz || framework_failure +lines=`"${LZIP}" -tvv out2.lz 2>&1 | wc -l` || test_failed $LINENO +[ "${lines}" -eq 2 ] || test_failed $LINENO "${lines}" +lines=`"${LZIP}" -lvv out2.lz | wc -l` || test_failed $LINENO +[ "${lines}" -eq 5 ] || test_failed $LINENO "${lines}" + printf "\ngarbage" >> out2.lz || framework_failure "${LZIP}" -tvvvv out2.lz 2> /dev/null || test_failed $LINENO -"${LZIPRECOVER}" -aD0 -q out2.lz -[ $? = 2 ] || test_failed $LINENO "${LZIP}" -alq out2.lz [ $? = 2 ] || test_failed $LINENO "${LZIP}" -atq out2.lz @@ -392,6 +378,8 @@ printf "\ngarbage" >> out2.lz || framework_failure "${LZIP}" -adkq -o out2 < out2.lz [ $? = 2 ] || test_failed $LINENO [ ! -e out2 ] || test_failed $LINENO +"${LZIPRECOVER}" -aD0 -q out2.lz +[ $? = 2 ] || test_failed $LINENO printf "to be overwritten" > out2 || framework_failure "${LZIP}" -df out2.lz || test_failed $LINENO cmp in2 out2 || test_failed $LINENO @@ -410,11 +398,21 @@ rm -rf a || framework_failure if [ -z "${LZIP_NAME}" ] ; then LZIP_NAME=lzip ; fi touch empty em || framework_failure "${LZIP_NAME}" -0 em || test_failed $LINENO -"${LZIP}" -lq em.lz || test_failed $LINENO -"${LZIP}" -d em.lz || test_failed $LINENO +"${LZIP}" -l em.lz > /dev/null || test_failed $LINENO +"${LZIP}" -dk em.lz || test_failed $LINENO cmp empty em || test_failed $LINENO -rm -f em || framework_failure +cat em.lz em.lz | "${LZIP}" -t || test_failed $LINENO +cat em.lz em.lz | "${LZIP}" -d > em || test_failed $LINENO +cmp empty em || test_failed $LINENO +cat em.lz "${in_lz}" | "${LZIP}" -t || test_failed $LINENO +cat em.lz "${in_lz}" | "${LZIP}" -d > out || test_failed $LINENO +cmp in out || test_failed $LINENO +cat "${in_lz}" em.lz | "${LZIP}" -t || test_failed $LINENO +cat "${in_lz}" em.lz | "${LZIP}" -d > out || test_failed $LINENO +cmp in out || test_failed $LINENO +cat em.lz "${fox_lz}" em.lz "${fox_lz}" em.lz em.lz "${fox_lz}" em.lz > \ + f3_em.lz || framework_failure "${LZIPRECOVER}" -D ,18000 "${in_lz}" > out || test_failed $LINENO "${LZIPRECOVER}" -D 18000 "${in_lz}" >> out || test_failed $LINENO cmp in out || test_failed $LINENO @@ -422,22 +420,60 @@ cmp in out || test_failed $LINENO cmp "${inD}" out || test_failed $LINENO "${LZIPRECOVER}" -D 21636,397 "${in_lz}" > out || test_failed $LINENO cmp "${inD}" out || test_failed $LINENO -"${LZIPRECOVER}" -D 21636,397 "${in_em}" --ignore-empty > out || - test_failed $LINENO -cmp "${inD}" out || test_failed $LINENO -"${LZIPRECOVER}" -q -D 21636,397 "${in_em}" -[ $? = 2 ] || test_failed $LINENO +"${LZIPRECOVER}" -D 45,45 f3_em.lz > out || test_failed $LINENO +cmp fox out || test_failed $LINENO "${LZIP}" -D 0 "${in_lz}" -o a/b/c/out || test_failed $LINENO cmp in a/b/c/out || test_failed $LINENO rm -rf a || framework_failure printf "\ntesting bad input..." +cat em.lz em.lz > ee.lz || framework_failure +"${LZIP}" -l < ee.lz > /dev/null || test_failed $LINENO +"${LZIP}" -t < ee.lz || test_failed $LINENO +"${LZIP}" -d < ee.lz > em || test_failed $LINENO +cmp empty em || test_failed $LINENO +"${LZIPRECOVER}" -li ee.lz > /dev/null || test_failed $LINENO +"${LZIPRECOVER}" -ti ee.lz || test_failed $LINENO +"${LZIPRECOVER}" -cdi ee.lz > em || test_failed $LINENO +cmp empty em || test_failed $LINENO +"${LZIP}" -lq ee.lz +[ $? = 2 ] || test_failed $LINENO +"${LZIP}" -tq ee.lz +[ $? = 2 ] || test_failed $LINENO +"${LZIP}" -dq ee.lz +[ $? = 2 ] || test_failed $LINENO +[ ! -e ee ] || test_failed $LINENO +"${LZIP}" -cdq ee.lz > em +[ $? = 2 ] || test_failed $LINENO +cmp empty em || test_failed $LINENO +rm -f em || framework_failure +cat "${in_lz}" em.lz "${in_lz}" > inein.lz || framework_failure +"${LZIP}" -l < inein.lz > /dev/null || test_failed $LINENO +"${LZIP}" -t < inein.lz || test_failed $LINENO +"${LZIP}" -d < inein.lz > out2 || test_failed $LINENO +cmp in2 out2 || test_failed $LINENO +"${LZIPRECOVER}" -li inein.lz > /dev/null || test_failed $LINENO +"${LZIPRECOVER}" -ti inein.lz || test_failed $LINENO +"${LZIPRECOVER}" -cdi inein.lz > out2 || test_failed $LINENO +cmp in2 out2 || test_failed $LINENO +"${LZIP}" -lq inein.lz +[ $? = 2 ] || test_failed $LINENO +"${LZIP}" -tq inein.lz +[ $? = 2 ] || test_failed $LINENO +"${LZIP}" -dq inein.lz +[ $? = 2 ] || test_failed $LINENO +[ ! -e inein ] || test_failed $LINENO +"${LZIP}" -cdq inein.lz > out2 +[ $? = 2 ] || test_failed $LINENO +cmp in2 out2 || test_failed $LINENO +rm -f out2 inein.lz em.lz || framework_failure + headers='LZIp LZiP LZip LzIP LzIp LziP lZIP lZIp lZiP lzIP' body='\001\014\000\000\101\376\367\377\377\340\000\200\000\215\357\002\322\001\000\000\000\000\000\000\000\045\000\000\000\000\000\000\000' cp "${in_lz}" int.lz || framework_failure printf "LZIP${body}" >> int.lz || framework_failure -if "${LZIP}" -tq int.lz ; then +if "${LZIP}" -t int.lz ; then for header in ${headers} ; do printf "${header}${body}" > int.lz || framework_failure "${LZIP}" -lq int.lz # first member @@ -466,7 +502,7 @@ if "${LZIP}" -tq int.lz ; then [ $? = 2 ] || test_failed $LINENO ${header} "${LZIP}" -cdq int.lz > /dev/null [ $? = 2 ] || test_failed $LINENO ${header} - "${LZIP}" -lq --loose-trailing int.lz || + "${LZIP}" -l --loose-trailing int.lz > /dev/null || test_failed $LINENO ${header} "${LZIP}" -t --loose-trailing int.lz || test_failed $LINENO ${header} @@ -497,24 +533,31 @@ if "${LZIP}" -tq int.lz ; then cmp "${in_lz}" int.lz || test_failed $LINENO ${header} done else - printf "\nwarning: skipping header test: 'printf' does not work on your system." + printf "warning: skipping header test: 'printf' does not work on your system." fi -rm -f int.lz out || framework_failure +rm -f int.lz || framework_failure -"${LZIP}" -cd "${fox6_lz}" > out || test_failed $LINENO -"${LZIPRECOVER}" -cd "${f6nz_lz}" --ignore-nonzero > copy || test_failed $LINENO -cmp copy out || test_failed $LINENO -rm -f copy out || framework_failure -"${LZIP}" -lq "${f6nz_lz}" || test_failed $LINENO -"${LZIP}" -tq "${f6nz_lz}" +"${LZIP}" -l "${fnz_lz}" > /dev/null || test_failed $LINENO +"${LZIP}" -tq "${fnz_lz}" [ $? = 2 ] || test_failed $LINENO -cp "${f6nz_lz}" f6nz.lz || framework_failure -cp "${f6nz_lz}" f6nz2.lz || framework_failure -cmp -s "${fox6_lz}" f6nz.lz && test_failed $LINENO -"${LZIPRECOVER}" --nonzero-repair f6nz.lz f6nz2.lz || test_failed $LINENO -cmp "${fox6_lz}" f6nz.lz || test_failed $LINENO -cmp "${fox6_lz}" f6nz2.lz || test_failed $LINENO -rm -f f6nz.lz f6nz2.lz || framework_failure +cat "${fox_lz}" "${fox_lz}" "${fox_lz}" "${fox_lz}" "${fox_lz}" "${fox_lz}" \ + > fox6.lz || framework_failure +cat "${fox_lz}" "${fnz_lz}" "${fox_lz}" "${fnz_lz}" "${fox_lz}" "${fnz_lz}" \ + > fox6_nz.lz || framework_failure +"${LZIP}" -cd fox6.lz > out || test_failed $LINENO +"${LZIPRECOVER}" -cd -i fox6_nz.lz > copy || test_failed $LINENO +cmp copy out || test_failed $LINENO +rm -f copy || framework_failure +cp fox6_nz.lz f6nz.lz || framework_failure +cp fox6_nz.lz f6nz2.lz || framework_failure +cp "${f6b1nz_lz}" f6b1nz.lz || framework_failure +cmp -s fox6.lz f6nz.lz && test_failed $LINENO +"${LZIPRECOVER}" --nonzero-repair f6nz.lz f6nz2.lz f6b1nz.lz || + test_failed $LINENO +cmp fox6.lz f6nz.lz || test_failed $LINENO +cmp fox6.lz f6nz2.lz || test_failed $LINENO +cmp "${f6b1_lz}" f6b1nz.lz || test_failed $LINENO +rm -f f6nz.lz f6nz2.lz f6b1nz.lz || framework_failure for i in fox_v2.lz fox_s11.lz fox_de20.lz \ fox_bcrc.lz fox_crc0.lz fox_das46.lz fox_mes81.lz ; do @@ -530,12 +573,12 @@ for i in fox_bcrc.lz fox_crc0.lz fox_das46.lz fox_mes81.lz ; do "${LZIPRECOVER}" -cdq -i "${testdir}"/$i > out || test_failed $LINENO $i cmp fox out || test_failed $LINENO $i done -rm -f fox out || framework_failure +rm -f fox || framework_failure cat "${in_lz}" "${in_lz}" > in2.lz || framework_failure cat "${in_lz}" "${in_lz}" "${in_lz}" > in3.lz || framework_failure if dd if=in3.lz of=trunc.lz bs=14682 count=1 2> /dev/null && - [ -e trunc.lz ] && cmp in2.lz trunc.lz > /dev/null 2>&1 ; then + [ -e trunc.lz ] && cmp in2.lz trunc.lz ; then for i in 6 20 14664 14683 14684 14685 14686 14687 14688 ; do dd if=in3.lz of=trunc.lz bs=$i count=1 2> /dev/null "${LZIP}" -lq trunc.lz @@ -550,7 +593,7 @@ if dd if=in3.lz of=trunc.lz bs=14682 count=1 2> /dev/null && [ $? = 2 ] || test_failed $LINENO $i done else - printf "\nwarning: skipping truncation test: 'dd' does not work on your system." + printf "warning: skipping truncation test: 'dd' does not work on your system." fi rm -f in3.lz trunc.lz || framework_failure @@ -572,17 +615,22 @@ cat "${in_lz}" >> ingin.lz || framework_failure [ $? = 2 ] || test_failed $LINENO "${LZIP}" -atq < ingin.lz [ $? = 2 ] || test_failed $LINENO -"${LZIP}" -acdq ingin.lz > /dev/null +"${LZIP}" -acdq ingin.lz > out [ $? = 2 ] || test_failed $LINENO -"${LZIP}" -adq < ingin.lz > /dev/null +cmp in out || test_failed $LINENO +"${LZIP}" -adq < ingin.lz > out [ $? = 2 ] || test_failed $LINENO -"${LZIPRECOVER}" -lq -i ingin.lz || test_failed $LINENO +cmp in out || test_failed $LINENO "${LZIP}" -t ingin.lz || test_failed $LINENO "${LZIP}" -t < ingin.lz || test_failed $LINENO +"${LZIP}" -dk ingin.lz || test_failed $LINENO +cmp in ingin || test_failed $LINENO "${LZIP}" -cd ingin.lz > out || test_failed $LINENO cmp in out || test_failed $LINENO "${LZIP}" -d < ingin.lz > out || test_failed $LINENO cmp in out || test_failed $LINENO +rm -f ingin || framework_failure +"${LZIPRECOVER}" -l -i ingin.lz > /dev/null || test_failed $LINENO "${LZIPRECOVER}" -cd -i ingin.lz > out2 || test_failed $LINENO cmp in2 out2 || test_failed $LINENO @@ -615,7 +663,7 @@ rm -f in2 in2t.lz out out2 || framework_failure printf "\ntesting --byte-repair..." rm -f out.lz || framework_failure -"${LZIPRECOVER}" -B -o out.lz "${fox6_lz}" || test_failed $LINENO +"${LZIPRECOVER}" -B -o out.lz fox6.lz || test_failed $LINENO [ ! -e out.lz ] || test_failed $LINENO "${LZIPRECOVER}" -B -o out.lz "${bad2_lz}" -q [ $? = 2 ] || test_failed $LINENO @@ -627,7 +675,11 @@ rm -f out.lz || framework_failure [ $? = 2 ] || test_failed $LINENO [ ! -e out.lz ] || test_failed $LINENO "${LZIPRECOVER}" -Bf -o out.lz "${f6b1_lz}" || test_failed $LINENO -cmp "${fox6_lz}" out.lz || test_failed $LINENO +cmp fox6.lz out.lz || test_failed $LINENO +"${LZIPRECOVER}" -Bf fox6_nz.lz -o out.lz || test_failed $LINENO +cmp fox6.lz out.lz || test_failed $LINENO +"${LZIPRECOVER}" -Bf "${f6b1nz_lz}" -o out.lz || test_failed $LINENO +cmp fox6.lz out.lz || test_failed $LINENO "${LZIPRECOVER}" -Bf -o out.lz "${bad1_lz}" || test_failed $LINENO cmp "${in_lz}" out.lz || test_failed $LINENO "${LZIPRECOVER}" -B -o a/b/c/out.lz "${bad1_lz}" || test_failed $LINENO @@ -667,21 +719,21 @@ cmp fec/test.txt.lz.fec fecfile.fec || test_failed $LINENO "${LZIPRECOVER}" -Ft "${in_lz}" --fec-file=fec/ || test_failed $LINENO rm -rf fec || framework_failure mkdir a -cp "${in_lz}" "${fox6_lz}" a || framework_failure +cp fox6.lz "${in_lz}" a || framework_failure "${LZIPRECOVER}" -r -Fc a/ -o fec/ || test_failed $LINENO -cmp fec/test.txt.lz.fec fecfile.fec || test_failed $LINENO [ -e fec/fox6.lz.fec ] || test_failed $LINENO +cmp fec/test.txt.lz.fec fecfile.fec || test_failed $LINENO "${LZIPRECOVER}" -r -Fc a -o fec/ || test_failed $LINENO -cmp fec/a/test.txt.lz.fec fecfile.fec || test_failed $LINENO [ -e fec/a/fox6.lz.fec ] || test_failed $LINENO +cmp fec/a/test.txt.lz.fec fecfile.fec || test_failed $LINENO rm -rf a fec fecfile.fec || framework_failure printf "\ntesting --merge..." rm -f out.lz || framework_failure -"${LZIPRECOVER}" -m -o out.lz "${fox6_lz}" "${f6b1_lz}" || test_failed $LINENO +"${LZIPRECOVER}" -m -o out.lz fox6.lz "${f6b1_lz}" || test_failed $LINENO [ ! -e out.lz ] || test_failed $LINENO -"${LZIPRECOVER}" -m -o out.lz "${f6b1_lz}" "${fox6_lz}" || test_failed $LINENO +"${LZIPRECOVER}" -m -o out.lz "${f6b1_lz}" fox6.lz || test_failed $LINENO [ ! -e out.lz ] || test_failed $LINENO "${LZIPRECOVER}" -m -o out.lz "${bad1_lz}" "${bad2_lz}" "${bad1_lz}" -q [ $? = 2 ] || test_failed $LINENO @@ -705,56 +757,56 @@ rm -f bad2.lz || framework_failure [ ! -e out.lz ] || test_failed $LINENO "${LZIPRECOVER}" -mf -o out.lz "${f6b1_lz}" "${f6b4_lz}" || test_failed $LINENO -cmp "${fox6_lz}" out.lz || test_failed $LINENO +cmp fox6.lz out.lz || test_failed $LINENO "${LZIPRECOVER}" -mf -o out.lz "${f6b4_lz}" "${f6b1_lz}" || test_failed $LINENO -cmp "${fox6_lz}" out.lz || test_failed $LINENO +cmp fox6.lz out.lz || test_failed $LINENO "${LZIPRECOVER}" -m -o a/b/c/out.lz "${f6b1_lz}" "${f6b4_lz}" || test_failed $LINENO -cmp "${fox6_lz}" a/b/c/out.lz || test_failed $LINENO +cmp fox6.lz a/b/c/out.lz || test_failed $LINENO rm -rf a || framework_failure for i in "${f6b1_lz}" "${f6b3_lz}" "${f6b4_lz}" "${f6b5_lz}" "${f6b6_lz}" ; do "${LZIPRECOVER}" -mf -o out.lz "${f6b2_lz}" "$i" || test_failed $LINENO "$i" - cmp "${fox6_lz}" out.lz || test_failed $LINENO "$i" + cmp fox6.lz out.lz || test_failed $LINENO "$i" "${LZIPRECOVER}" -mf -o out.lz "$i" "${f6b2_lz}" || test_failed $LINENO "$i" - cmp "${fox6_lz}" out.lz || test_failed $LINENO "$i" + cmp fox6.lz out.lz || test_failed $LINENO "$i" done for i in "${f6b3_lz}" "${f6b4_lz}" "${f6b5_lz}" "${f6b6_lz}" ; do "${LZIPRECOVER}" -mf -o out.lz "${f6b1_lz}" "${f6b2_lz}" "$i" || test_failed $LINENO "$i" - cmp "${fox6_lz}" out.lz || test_failed $LINENO "$i" + cmp fox6.lz out.lz || test_failed $LINENO "$i" "${LZIPRECOVER}" -mf -o out.lz "${f6b1_lz}" "$i" "${f6b2_lz}" || test_failed $LINENO "$i" - cmp "${fox6_lz}" out.lz || test_failed $LINENO "$i" + cmp fox6.lz out.lz || test_failed $LINENO "$i" "${LZIPRECOVER}" -mf -o out.lz "${f6b2_lz}" "${f6b1_lz}" "$i" || test_failed $LINENO "$i" - cmp "${fox6_lz}" out.lz || test_failed $LINENO "$i" + cmp fox6.lz out.lz || test_failed $LINENO "$i" "${LZIPRECOVER}" -mf -o out.lz "${f6b2_lz}" "$i" "${f6b1_lz}" || test_failed $LINENO "$i" - cmp "${fox6_lz}" out.lz || test_failed $LINENO "$i" + cmp fox6.lz out.lz || test_failed $LINENO "$i" "${LZIPRECOVER}" -mf -o out.lz "$i" "${f6b1_lz}" "${f6b2_lz}" || test_failed $LINENO "$i" - cmp "${fox6_lz}" out.lz || test_failed $LINENO "$i" + cmp fox6.lz out.lz || test_failed $LINENO "$i" "${LZIPRECOVER}" -mf -o out.lz "$i" "${f6b2_lz}" "${f6b1_lz}" || test_failed $LINENO "$i" - cmp "${fox6_lz}" out.lz || test_failed $LINENO "$i" + cmp fox6.lz out.lz || test_failed $LINENO "$i" done "${LZIPRECOVER}" -mf -o out.lz "${f6b3_lz}" "${f6b4_lz}" "${f6b5_lz}" || test_failed $LINENO -cmp "${fox6_lz}" out.lz || test_failed $LINENO +cmp fox6.lz out.lz || test_failed $LINENO "${LZIPRECOVER}" -mf -o out.lz "${f6b1_lz}" "${f6b3_lz}" "${f6b4_lz}" \ "${f6b5_lz}" || test_failed $LINENO -cmp "${fox6_lz}" out.lz || test_failed $LINENO +cmp fox6.lz out.lz || test_failed $LINENO "${LZIPRECOVER}" -mf -o out.lz "${f6b2_lz}" "${f6b3_lz}" "${f6b4_lz}" \ "${f6b5_lz}" || test_failed $LINENO -cmp "${fox6_lz}" out.lz || test_failed $LINENO +cmp fox6.lz out.lz || test_failed $LINENO "${LZIPRECOVER}" -mf -o out.lz "${f6b1_lz}" "${f6b2_lz}" "${f6b3_lz}" \ "${f6b4_lz}" "${f6b5_lz}" || test_failed $LINENO -cmp "${fox6_lz}" out.lz || test_failed $LINENO +cmp fox6.lz out.lz || test_failed $LINENO "${LZIPRECOVER}" -mf -o out.lz "${bad1_lz}" "${bad2_lz}" || test_failed $LINENO cmp "${in_lz}" out.lz || test_failed $LINENO @@ -836,7 +888,7 @@ rm -f bad345.lz bad453.lz bad534.lz out4.lz || framework_failure printf "\ntesting --reproduce..." if /bin/sh -c "${LZIP_NAME} -s18KiB" < in > out 2> /dev/null && - cmp "${in_lz}" out > /dev/null 2>&1 ; then + cmp "${in_lz}" out ; then rm -f out || framework_failure "${LZIPRECOVER}" --reproduce --lzip-name="${LZIP_NAME}" -o out \ --reference-file=foo "${in_lz}" || test_failed $LINENO "${LZIP_NAME}" @@ -850,19 +902,19 @@ if /bin/sh -c "${LZIP_NAME} -s18KiB" < in > out 2> /dev/null && for i in 6 7 8 9 ; do for f in "${testdir}"/test_bad${i}.txt in ; do rm -f out || framework_failure - "${LZIPRECOVER}" -q --reproduce --lzip-name="${LZIP_NAME}" \ - --reference-file="$f" "${testdir}"/test_bad${i}.lz -o out || + "${LZIPRECOVER}" --reproduce --lzip-name="${LZIP_NAME}" -o out \ + --reference-file="$f" "${testdir}"/test_bad${i}.lz > /dev/null || test_failed $LINENO "${LZIP_NAME} $i $f" cmp "${in_lz}" out || test_failed $LINENO "${LZIP_NAME} $i $f" rm -f out || framework_failure - "${LZIPRECOVER}" -q --reproduce --lzip-name="${LZIP_NAME}" \ - --reference-file="$f" "${testdir}"/test_bad${i}.lz -o out \ - --lzip-level=6 || test_failed $LINENO "${LZIP_NAME} $i $f level=6" + "${LZIPRECOVER}" --reproduce --lzip-name="${LZIP_NAME}" -o out \ + --lzip-level=6 --reference-file="$f" "${testdir}"/test_bad${i}.lz \ + > /dev/null || test_failed $LINENO "${LZIP_NAME} $i $f level=6" cmp "${in_lz}" out || test_failed $LINENO "${LZIP_NAME} $i $f level=6" rm -f out || framework_failure - "${LZIPRECOVER}" -q --reproduce --lzip-name="${LZIP_NAME}" \ - --reference-file="$f" "${testdir}"/test_bad${i}.lz -o out \ - --lzip-level=m36 || test_failed $LINENO "${LZIP_NAME} $i $f level=m36" + "${LZIPRECOVER}" --reproduce --lzip-name="${LZIP_NAME}" -o out \ + --lzip-level=m36 --reference-file="$f" "${testdir}"/test_bad${i}.lz \ + > /dev/null || test_failed $LINENO "${LZIP_NAME} $i $f level=m36" cmp "${in_lz}" out || test_failed $LINENO "${LZIP_NAME} $i $f level=m36" done done @@ -872,8 +924,8 @@ if /bin/sh -c "${LZIP_NAME} -s18KiB" < in > out 2> /dev/null && framework_failure rm -f out || framework_failure for i in 6 7 8 9 ; do # reproduce one member each time - "${LZIPRECOVER}" -q --reproduce --lzip-name="${LZIP_NAME}" \ - --reference-file="${testdir}"/test_bad${i}.txt mm_bad.lz -o out || + "${LZIPRECOVER}" --reproduce --lzip-name="${LZIP_NAME}" mm_bad.lz \ + --reference-file="${testdir}"/test_bad${i}.txt -o out > /dev/null || test_failed $LINENO "${LZIP_NAME} $i" mv out mm_bad.lz || framework_failure done @@ -884,22 +936,23 @@ if /bin/sh -c "${LZIP_NAME} -s18KiB" < in > out 2> /dev/null && framework_failure rm -f out || framework_failure for i in 6 7 8 9 ; do # reproduce one member each time - "${LZIPRECOVER}" -q --reproduce --lzip-name="${LZIP_NAME}" -o out \ - --reference-file=in mm_bad.lz || test_failed $LINENO "${LZIP_NAME} $i" + "${LZIPRECOVER}" --reproduce --lzip-name="${LZIP_NAME}" mm_bad.lz -o out \ + --reference-file=in > /dev/null || test_failed $LINENO "${LZIP_NAME} $i" mv out mm_bad.lz || framework_failure done cmp in4.lz mm_bad.lz || test_failed $LINENO "${LZIP_NAME}" rm -f mm_bad.lz || framework_failure - "${LZIPRECOVER}" -q --debug-reproduce=13-7356 --lzip-name="${LZIP_NAME}" \ - --reference-file=in "${in_lz}" || test_failed $LINENO "${LZIP_NAME}" + "${LZIPRECOVER}" --reference-file=in "${in_lz}" --lzip-name="${LZIP_NAME}" \ + --debug-reproduce=13-7356 > /dev/null || test_failed $LINENO "${LZIP_NAME}" "${LZIPRECOVER}" --debug-reproduce=512,5120,512 --lzip-name="${LZIP_NAME}" \ -q --reference-file=in "${in_lz}" || test_failed $LINENO "${LZIP_NAME}" else - printf "\nwarning: skipping --reproduce test: ${LZIP_NAME} not found or not the right version.\n" - ${LZIP_NAME} -V - printf "\nTry 'make LZIP_NAME=<name_of_lzip_executable> check'." + printf "warning: skipping --reproduce test: " + printf "${LZIP_NAME} not found, not the right compressor, not the right version, or this is not a POSIX system.\n" + if ${LZIP_NAME} -V > /dev/null 2>&1 ; then ${LZIP_NAME} -V | sed -e 1q + else printf "Try 'make LZIP_NAME=<name_of_lzip_executable> check'.\n" ; fi fi rm -f in4.lz || framework_failure @@ -1222,21 +1275,17 @@ for i in 1 2 3 4 5 6 7 8 9 10 ; do done rm -f nbt.lz || framework_failure -cp "${in_em}" test_3m.txt.lz || framework_failure -"${LZIPRECOVER}" --remove=empty test_3m.txt.lz || test_failed $LINENO -"${LZIPRECOVER}" -M test_3m.txt.lz | cmp "${testdir}"/test_3m.txt.lz.md5 - || - test_failed $LINENO -"${LZIPRECOVER}" --dump=2,4,7 "${in_em}" | cmp test_3m.txt.lz - || - test_failed $LINENO -"${LZIPRECOVER}" --strip=e "${in_em}" | cmp test_3m.txt.lz - || - test_failed $LINENO -"${LZIPRECOVER}" --strip=e "${in_em}" | cmp test_3m.txt.lz - || +cat "${fox_lz}" "${fox_lz}" "${fox_lz}" > fox3.lz || framework_failure +cp f3_em.lz out.lz || framework_failure +"${LZIPRECOVER}" --remove=empty out.lz || test_failed $LINENO +cmp fox3.lz out.lz || test_failed $LINENO +rm -f out.lz || framework_failure +"${LZIPRECOVER}" --dump=2,4,7 f3_em.lz | cmp fox3.lz - || test_failed $LINENO +"${LZIPRECOVER}" --strip=e f3_em.lz | cmp fox3.lz - || test_failed $LINENO +"${LZIPRECOVER}" --strip=1,3,5-6,8 f3_em.lz | cmp fox3.lz - || test_failed $LINENO -"${LZIPRECOVER}" --strip=1,3,5-6,8 "${in_em}" | cmp test_3m.txt.lz - || +"${LZIPRECOVER}" --dump=empty f3_em.lz | "${LZIP}" -d | cmp empty - || test_failed $LINENO -"${LZIPRECOVER}" --dump=emp "${in_em}" | "${LZIP}" -d --ignore-empty | \ - cmp empty - || test_failed $LINENO -rm -f test_3m.txt.lz || framework_failure printf "\ntesting --dump/remove/strip=damaged..." @@ -1295,9 +1344,9 @@ cmp "${f6b1_lz}" out || test_failed $LINENO cat "${f6b1_lz}" in > f6bt.lz || framework_failure "${LZIPRECOVER}" --dump=damaged f6bt.lz > out || test_failed $LINENO cmp "${f6b1_lz}" out || test_failed $LINENO -"${LZIPRECOVER}" -q --strip=damaged "${f6b1_lz}" > out || test_failed $LINENO +"${LZIPRECOVER}" --strip=damaged "${f6b1_lz}" > out || test_failed $LINENO cmp empty out || test_failed $LINENO -"${LZIPRECOVER}" -q --strip=damaged f6bt.lz > out || test_failed $LINENO +"${LZIPRECOVER}" --strip=damaged f6bt.lz > out || test_failed $LINENO cmp empty out || test_failed $LINENO cp "${f6b1_lz}" f6b.lz || framework_failure "${LZIPRECOVER}" -q --remove=damaged f6b.lz @@ -1335,7 +1384,6 @@ cat "${f6b3_lz}" in > f6bt.lz || framework_failure "${LZIPRECOVER}" --dump=damaged f6bt.lz > out || test_failed $LINENO cat "${fox_lz}" "${fox_lz}" out "${fox_lz}" | cmp "${f6b3_lz}" - || test_failed $LINENO -cat "${fox_lz}" "${fox_lz}" "${fox_lz}" > fox3.lz || framework_failure "${LZIPRECOVER}" --strip=damaged "${f6b3_lz}" > out || test_failed $LINENO cmp fox3.lz out || test_failed $LINENO "${LZIPRECOVER}" --strip=damaged f6bt.lz > out || test_failed $LINENO @@ -1392,10 +1440,9 @@ for i in "${f6s1_lz}" "${f6s2_lz}" ; do "${LZIPRECOVER}" --dump=damaged f6bt.lz > out || test_failed $LINENO "$i" cmp "$i" out || test_failed $LINENO "$i" - "${LZIPRECOVER}" -q --strip=damaged "$i" > out || - test_failed $LINENO "$i" + "${LZIPRECOVER}" --strip=damaged "$i" > out || test_failed $LINENO "$i" cmp empty out || test_failed $LINENO "$i" - "${LZIPRECOVER}" -q --strip=damaged f6bt.lz > out || + "${LZIPRECOVER}" --strip=damaged f6bt.lz > out || test_failed $LINENO "$i" cmp empty out || test_failed $LINENO "$i" cp "$i" f6b.lz || framework_failure @@ -1458,22 +1505,22 @@ cat "${bad2_lz}" out "${bad2_lz}" out | cmp out4 - || test_failed $LINENO "${bad2_lz}" > out4 || test_failed $LINENO cat out "${bad2_lz}" out "${bad2_lz}" | cmp out4 - || test_failed $LINENO # -"${LZIPRECOVER}" -q --strip=damaged "${bad2_lz}" "${f6b2_lz}" > out || +"${LZIPRECOVER}" --strip=damaged "${bad2_lz}" "${f6b2_lz}" > out || test_failed $LINENO cmp fox5.lz out || test_failed $LINENO -"${LZIPRECOVER}" -q --strip=damaged bad2t.lz "${f6b2_lz}" > out || +"${LZIPRECOVER}" --strip=damaged bad2t.lz "${f6b2_lz}" > out || test_failed $LINENO cmp fox5.lz out || test_failed $LINENO -"${LZIPRECOVER}" -q --strip=damaged "${f6b2_lz}" bad2t.lz f6bt.lz > out || +"${LZIPRECOVER}" --strip=damaged "${f6b2_lz}" bad2t.lz f6bt.lz > out || test_failed $LINENO cat fox5.lz fox5.lz in | cmp out - || test_failed $LINENO -"${LZIPRECOVER}" -q --strip=damaged "${f6b2_lz}" f6bt.lz bad2t.lz > out || +"${LZIPRECOVER}" --strip=damaged "${f6b2_lz}" f6bt.lz bad2t.lz > out || test_failed $LINENO cat fox5.lz fox5.lz | cmp out - || test_failed $LINENO -"${LZIPRECOVER}" -q --strip=damaged f6bt.lz bad2t.lz > out || +"${LZIPRECOVER}" --strip=damaged f6bt.lz bad2t.lz > out || test_failed $LINENO cmp fox5.lz out || test_failed $LINENO -"${LZIPRECOVER}" -q --strip=damaged f6bt.lz "${in_lz}" > out || +"${LZIPRECOVER}" --strip=damaged f6bt.lz "${in_lz}" > out || test_failed $LINENO cat fox5.lz "${in_lz}" | cmp out - || test_failed $LINENO "${LZIPRECOVER}" --strip=damaged --strip=tdata f6bt.lz "${in_lz}" > out || @@ -1487,7 +1534,7 @@ cat "${bad2_lz}" in | cmp bad2t.lz - || test_failed $LINENO cmp fox5.lz f6b.lz || test_failed $LINENO cat fox5.lz in | cmp f6bt.lz - || test_failed $LINENO cat "${bad2_lz}" in > bad2t.lz || framework_failure -cat "${fox6_lz}" "${inD}" > fox6t.lz || framework_failure +cat fox6.lz "${inD}" > fox6t.lz || framework_failure cat "${f6b1_lz}" in > f6abt.lz || framework_failure cp "${f6b2_lz}" f6b.lz || framework_failure cat "${f6b2_lz}" in > f6bt.lz || framework_failure @@ -1495,7 +1542,7 @@ cat "${f6b2_lz}" in > f6bt.lz || framework_failure [ $? = 2 ] || test_failed $LINENO cat "${bad2_lz}" in | cmp bad2t.lz - || test_failed $LINENO cat "${f6b1_lz}" in | cmp f6abt.lz - || test_failed $LINENO -cmp "${fox6_lz}" fox6t.lz || test_failed $LINENO +cmp fox6.lz fox6t.lz || test_failed $LINENO cmp fox5.lz f6b.lz || test_failed $LINENO cmp fox5.lz f6bt.lz || test_failed $LINENO rm -f fox6t.lz f6b.lz f6bt.lz f6abt.lz bad2t.lz fox5.lz out2 out4 || @@ -1509,7 +1556,7 @@ cmp "${inD}" out || test_failed $LINENO rm -f out || framework_failure "${LZIPRECOVER}" --dump=tdat int.lz -o out || test_failed $LINENO cmp "${inD}" out || test_failed $LINENO -cat "${fox6_lz}" "${inD}" > fox6t.lz || framework_failure +cat fox6.lz "${inD}" > fox6t.lz || framework_failure cat "${inD}" "${inD}" > inD2 || framework_failure "${LZIPRECOVER}" --dump=tda int.lz fox6t.lz -f -o out || test_failed $LINENO cmp inD2 out || test_failed $LINENO @@ -1526,12 +1573,12 @@ rm -f out || framework_failure "${LZIPRECOVER}" --strip=tdata int.lz -o out || test_failed $LINENO cmp "${in_lz}" out || test_failed $LINENO "${LZIPRECOVER}" --strip=tdata fox6t.lz -f -o out || test_failed $LINENO -cmp "${fox6_lz}" out || test_failed $LINENO +cmp fox6.lz out || test_failed $LINENO "${LZIPRECOVER}" --strip=tdata int.lz int.lz -f -o out || test_failed $LINENO cmp in2.lz out || test_failed $LINENO rm -f in2.lz || framework_failure "${LZIPRECOVER}" --strip=tdata int.lz fox6t.lz > out || test_failed $LINENO -cat "${in_lz}" "${fox6_lz}" | cmp out - || test_failed $LINENO +cat "${in_lz}" fox6.lz | cmp out - || test_failed $LINENO "${LZIPRECOVER}" -q --strip=tdata ingint.lz > out # /dev/null returns 1 on OS/2 [ $? = 2 ] || test_failed $LINENO "${LZIPRECOVER}" -i --strip=tdata ingint.lz > out || test_failed $LINENO @@ -1539,11 +1586,11 @@ cmp ingin.lz out || test_failed $LINENO "${LZIPRECOVER}" --remove=tdata int.lz fox6t.lz || test_failed $LINENO cmp "${in_lz}" int.lz || test_failed $LINENO -cmp "${fox6_lz}" fox6t.lz || test_failed $LINENO +cmp fox6.lz fox6t.lz || test_failed $LINENO "${LZIPRECOVER}" --remove=tdata int.lz || test_failed $LINENO cmp "${in_lz}" int.lz || test_failed $LINENO "${LZIPRECOVER}" --remove=tdata fox6t.lz || test_failed $LINENO -cmp "${fox6_lz}" fox6t.lz || test_failed $LINENO +cmp fox6.lz fox6t.lz || test_failed $LINENO "${LZIPRECOVER}" -q --remove=tdata ingint.lz [ $? = 2 ] || test_failed $LINENO cmp -s ingin.lz ingint.lz && test_failed $LINENO @@ -1554,15 +1601,15 @@ rm -f int.lz fox6t.lz ingint.lz ingin.lz || framework_failure for i in "${f6s3_lz}" "${f6s4_lz}" "${f6s5_lz}" "${f6s6_lz}" ; do "${LZIPRECOVER}" --strip=tdata "$i" > out || test_failed $LINENO "$i" "${LZIPRECOVER}" --dump=tdata "$i" > tdata || test_failed $LINENO "$i" - cmp "${fox6_lz}" out || test_failed $LINENO "$i" + cmp fox6.lz out || test_failed $LINENO "$i" cat out tdata | cmp "$i" - || test_failed $LINENO "$i" cat "$i" "${inD}" > f6t.lz || framework_failure "${LZIPRECOVER}" --strip=tdata f6t.lz > out || test_failed $LINENO "$i" "${LZIPRECOVER}" --dump=tdata f6t.lz > tdata || test_failed $LINENO "$i" - cmp "${fox6_lz}" out || test_failed $LINENO "$i" + cmp fox6.lz out || test_failed $LINENO "$i" cat out tdata | cmp f6t.lz - || test_failed $LINENO "$i" "${LZIPRECOVER}" --remove=tdata f6t.lz || test_failed $LINENO "$i" - cmp "${fox6_lz}" f6t.lz || test_failed $LINENO "$i" + cmp fox6.lz f6t.lz || test_failed $LINENO "$i" rm -f out tdata f6t.lz || framework_failure done diff --git a/testsuite/fox6.lz b/testsuite/fox6.lz Binary files differdeleted file mode 100644 index 8401b99..0000000 --- a/testsuite/fox6.lz +++ /dev/null diff --git a/testsuite/fox6_b1nz.lz b/testsuite/fox6_b1nz.lz Binary files differnew file mode 100644 index 0000000..d841502 --- /dev/null +++ b/testsuite/fox6_b1nz.lz diff --git a/testsuite/fox6_nz.lz b/testsuite/fox6_nz.lz Binary files differdeleted file mode 100644 index cd5b481..0000000 --- a/testsuite/fox6_nz.lz +++ /dev/null diff --git a/testsuite/fox_nz.lz b/testsuite/fox_nz.lz Binary files differnew file mode 100644 index 0000000..44a4b58 --- /dev/null +++ b/testsuite/fox_nz.lz diff --git a/testsuite/test_3m.txt.lz.md5 b/testsuite/test_3m.txt.lz.md5 deleted file mode 100644 index e9f47d7..0000000 --- a/testsuite/test_3m.txt.lz.md5 +++ /dev/null @@ -1 +0,0 @@ -aa8ca65001d627f89e7494fa829e710f test_3m.txt.lz diff --git a/testsuite/test_em.txt.lz b/testsuite/test_em.txt.lz Binary files differdeleted file mode 100644 index 9e093be..0000000 --- a/testsuite/test_em.txt.lz +++ /dev/null diff --git a/unzcrash.cc b/unzcrash.cc index b0c4871..d0e5049 100644 --- a/unzcrash.cc +++ b/unzcrash.cc @@ -149,7 +149,7 @@ uint8_t * read_file( const char * const filename, long * const file_sizep ) if( buffer_size >= LONG_MAX ) { show_file_error( filename, large_file_msg ); std::free( buffer ); return 0; } - buffer_size = ( buffer_size <= LONG_MAX / 2 ) ? 2 * buffer_size : LONG_MAX; + buffer_size = (buffer_size <= LONG_MAX / 2) ? 2 * buffer_size : LONG_MAX; uint8_t * const tmp = (uint8_t *)std::realloc( buffer, buffer_size ); if( !tmp ) { show_file_error( filename, mem_msg ); std::free( buffer ); return 0; } @@ -354,7 +354,7 @@ int main( const int argc, const char * const argv[] ) { enum Mode { m_block, m_byte, m_truncate }; const char * mode_str[3] = { "block", "byte", "size" }; - Bitset8 bits; // if Bitset8::parse_bs not called test full byte + Bitset8 bits; // if Bitset8::parse_bs not called, test full byte Bad_byte bad_byte; const char * zcmp_program = "zcmp"; long pos = 0; @@ -423,7 +423,7 @@ int main( const int argc, const char * const argv[] ) return 1; } - if( delta <= 0 ) delta = ( program_mode == m_block ) ? block_size : 1; + if( delta <= 0 ) delta = (program_mode == m_block) ? block_size : 1; const char * const command = parser.argument( argind ).c_str(); std::vector< std::string > command_args; @@ -498,7 +498,7 @@ int main( const int argc, const char * const argv[] ) ( max_size < 0 && -max_size >= file_size - pos ) ) { show_error( "Nothing to do; domain is empty." ); return 0; } if( max_size < 0 ) max_size += file_size - pos; - const long end = ( ( max_size < file_size - pos ) ? pos + max_size : file_size ); + const long end = (max_size < file_size - pos) ? pos + max_size : file_size; if( bad_byte.pos >= file_size ) { show_option_error( bad_byte.argument, "Position is beyond end of file in", bad_byte.option_name ); return 1; } |