From f8ec1a9b6de5ccde4220017676cbae1c16205fd1 Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Sat, 7 Nov 2015 11:16:13 +0100 Subject: Merging upstream version 1.13~rc2. Signed-off-by: Daniel Baumann --- ChangeLog | 11 ++ Makefile.in | 4 +- NEWS | 7 + configure | 2 +- decoder.cc | 11 +- decoder.h | 9 +- doc/lziprecover.1 | 13 +- doc/lziprecover.info | 46 +++++- doc/lziprecover.texinfo | 41 +++++- lzip.h | 40 +++++ main.cc | 143 ++++++++++-------- merge.cc | 25 +--- range_dec.cc | 359 +++++++++++++++++++++++++++++++++++++++++++++ repair.cc | 3 +- split.cc | 1 + testsuite/check.sh | 73 +++++---- testsuite/test921-1921.txt | 17 +++ 17 files changed, 676 insertions(+), 129 deletions(-) create mode 100644 range_dec.cc create mode 100644 testsuite/test921-1921.txt diff --git a/ChangeLog b/ChangeLog index 15e7746..8020d19 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,14 @@ +2011-11-20 Antonio Diaz Diaz + + * Version 1.13-rc2 released. + * Added new option `-D, --range-decompress' which extracts a + range of bytes decompressing only the members containing the + desired data. + * Added new option `-l, --list' which prints correct total file + sizes and ratios even for multimember files. + * New file range_dec.cc. + * testsuite/check.sh: Do not use `ln'. `ln' doesn't work on OS/2. + 2011-11-12 Antonio Diaz Diaz * Version 1.13-rc1 released. diff --git a/Makefile.in b/Makefile.in index 6c12f88..cc8d1c9 100644 --- a/Makefile.in +++ b/Makefile.in @@ -6,7 +6,7 @@ INSTALL_DATA = $(INSTALL) -p -m 644 INSTALL_DIR = $(INSTALL) -d -m 755 SHELL = /bin/sh -objs = arg_parser.o decoder.o merge.o repair.o split.o main.o +objs = arg_parser.o decoder.o merge.o range_dec.o repair.o split.o main.o .PHONY : all install install-info install-man install-strip \ @@ -32,6 +32,7 @@ arg_parser.o : arg_parser.h decoder.o : lzip.h decoder.h main.o : arg_parser.h lzip.h decoder.h merge.o : lzip.h decoder.h +range_dec.o : lzip.h decoder.h repair.o : lzip.h split.o : lzip.h @@ -97,6 +98,7 @@ dist : doc $(DISTNAME)/doc/$(pkgname).texinfo \ $(DISTNAME)/testsuite/check.sh \ $(DISTNAME)/testsuite/test.txt \ + $(DISTNAME)/testsuite/test921-1921.txt \ $(DISTNAME)/testsuite/test_bad[1-5].lz \ $(DISTNAME)/testsuite/test_v[01].lz \ $(DISTNAME)/*.h \ diff --git a/NEWS b/NEWS index 12d88ed..d8bb1ed 100644 --- a/NEWS +++ b/NEWS @@ -7,4 +7,11 @@ Decompressor options (-c, -d, -k, -t) have been implemented in lziprecover so that a external decompressor is not needed for recovery nor for "make check". +The new option "-D, --range-decompress" which extracts a range of bytes +decompressing only the members containing the desired data, has been +added. + +The new option "-l, --list" which prints correct total file sizes and +ratios even for multimember files, has been added. + "--merge" and "--repair" now remove the output file if recovery fails. diff --git a/configure b/configure index 43bf621..6fab087 100755 --- a/configure +++ b/configure @@ -8,7 +8,7 @@ args= no_create= pkgname=lziprecover -pkgversion=1.13-rc1 +pkgversion=1.13-rc2 progname=lziprecover srctrigger=lzip.h diff --git a/decoder.cc b/decoder.cc index b9fc3d8..e15b2ec 100644 --- a/decoder.cc +++ b/decoder.cc @@ -108,9 +108,14 @@ void LZ_decoder::flush_data() if( size > 0 ) { crc32.update( crc_, buffer + stream_pos, size ); - if( outfd >= 0 && - writeblock( outfd, buffer + stream_pos, size ) != size ) - throw Error( "Write error" ); + if( outfd >= 0 ) + { + const long long i = std::max( 0LL, outskip - stream_position() ); + const long long s = + std::min( outend - stream_position(), (long long)size ) - i; + if( s > 0 && writeblock( outfd, buffer + stream_pos + i, s ) != s ) + throw Error( "Write error" ); + } if( pos >= buffer_size ) { partial_data_pos += pos; pos = 0; } stream_pos = pos; } diff --git a/decoder.h b/decoder.h index 3b5dc7d..f1a48cb 100644 --- a/decoder.h +++ b/decoder.h @@ -192,6 +192,8 @@ public: class LZ_decoder { + const long long outskip; + const long long outend; long long partial_data_pos; const int dictionary_size; const int buffer_size; @@ -203,6 +205,8 @@ class LZ_decoder const int member_version; Range_decoder & range_decoder; + long long stream_position() const throw() + { return partial_data_pos + stream_pos; } void flush_data(); bool verify_trailer( const Pretty_print & pp ) const; @@ -243,8 +247,11 @@ class LZ_decoder } public: - LZ_decoder( const File_header & header, Range_decoder & rdec, const int ofd ) + LZ_decoder( const File_header & header, Range_decoder & rdec, const int ofd, + const long long oskip = 0, const long long oend = LLONG_MAX ) : + outskip( oskip ), + outend( oend ), partial_data_pos( 0 ), dictionary_size( header.dictionary_size() ), buffer_size( std::max( 65536, dictionary_size ) ), diff --git a/doc/lziprecover.1 b/doc/lziprecover.1 index 495de0b..f8e6148 100644 --- a/doc/lziprecover.1 +++ b/doc/lziprecover.1 @@ -1,5 +1,5 @@ .\" DO NOT MODIFY THIS FILE! It was generated by help2man 1.37.1. -.TH LZIPRECOVER "1" "November 2011" "Lziprecover 1.13-rc1" "User Commands" +.TH LZIPRECOVER "1" "November 2011" "Lziprecover 1.13-rc2" "User Commands" .SH NAME Lziprecover \- recovers data from damaged lzip files .SH SYNOPSIS @@ -21,12 +21,18 @@ send decompressed output to standard output \fB\-d\fR, \fB\-\-decompress\fR decompress .TP +\fB\-D\fR, \fB\-\-range\-decompress=\fR +decompress only a range of bytes (N\-M) +.TP \fB\-f\fR, \fB\-\-force\fR overwrite existing output files .TP \fB\-k\fR, \fB\-\-keep\fR keep (don't delete) input files .TP +\fB\-l\fR, \fB\-\-list\fR +print total file sizes and ratios +.TP \fB\-m\fR, \fB\-\-merge\fR correct errors in file using several copies .TP @@ -40,13 +46,16 @@ suppress all messages try to repair a small error in file .TP \fB\-s\fR, \fB\-\-split\fR -split a multimember file in single\-member files +split multimember file in single\-member files .TP \fB\-t\fR, \fB\-\-test\fR test compressed file integrity .TP \fB\-v\fR, \fB\-\-verbose\fR be verbose (a 2nd \fB\-v\fR gives more) +.PP +Numbers may be followed by a multiplier: k = kB = 10^3 = 1000, +Ki = KiB = 2^10 = 1024, M = 10^6, Mi = 2^20, G = 10^9, Gi = 2^30, etc... .SH "REPORTING BUGS" Report bugs to lzip\-bug@nongnu.org .br diff --git a/doc/lziprecover.info b/doc/lziprecover.info index 4528103..621f5cf 100644 --- a/doc/lziprecover.info +++ b/doc/lziprecover.info @@ -12,7 +12,7 @@ File: lziprecover.info, Node: Top, Next: Introduction, Up: (dir) Lziprecover Manual ****************** -This manual is for Lziprecover (version 1.13-rc1, 12 November 2011). +This manual is for Lziprecover (version 1.13-rc2, 20 November 2011). * Menu: @@ -101,6 +101,19 @@ The format for running lziprecover is: `--decompress' Decompress. +`-D RANGE' +`--range-decompress=RANGE' + Decompress only a range of bytes starting at decompressed byte + position `BEGIN' and up to byte position `END - 1'. Three formats + of RANGE are recognized, `BEGIN', `BEGIN-END', and `BEGIN,SIZE'. + If only BEGIN is specified, END is taken as the end of the file. + The produced bytes are sent to standard output unless the + `--output' option is used. 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 verified. + This operation is more efficient in multimember files because it + only decompresses the members containing the desired data. + `-f' `--force' Force overwrite of output files. @@ -109,6 +122,11 @@ The format for running lziprecover is: `--keep' Keep (don't delete) input files during decompression. +`-l' +`--list' + Print total file sizes and ratios. The values produced are correct + even for multimember files. + `-m' `--merge' Try to produce a correct file merging the good parts of two or more @@ -174,6 +192,22 @@ The format for running lziprecover is: bytes of trailing garbage (if any). + + Numbers given as arguments to options may be followed by a multiplier +and an optional `B' for "byte". + + Table of SI and binary prefixes (unit multipliers): + +Prefix Value | Prefix Value +k kilobyte (10^3 = 1000) | Ki kibibyte (2^10 = 1024) +M megabyte (10^6) | Mi mebibyte (2^20) +G gigabyte (10^9) | Gi gibibyte (2^30) +T terabyte (10^12) | Ti tebibyte (2^40) +P petabyte (10^15) | Pi pebibyte (2^50) +E exabyte (10^18) | Ei exbibyte (2^60) +Z zettabyte (10^21) | Zi zebibyte (2^70) +Y yottabyte (10^24) | Yi yobibyte (2^80) +  File: lziprecover.info, Node: File Format, Next: Examples, Prev: Invoking Lziprecover, Up: Top @@ -359,10 +393,10 @@ Tag Table: Node: Top231 Node: Introduction898 Node: Invoking Lziprecover2684 -Node: File Format6150 -Node: Examples8156 -Ref: ddrescue-example9409 -Node: Problems11188 -Node: Concept Index11738 +Node: File Format7727 +Node: Examples9733 +Ref: ddrescue-example10986 +Node: Problems12765 +Node: Concept Index13315  End Tag Table diff --git a/doc/lziprecover.texinfo b/doc/lziprecover.texinfo index 2447c14..9086787 100644 --- a/doc/lziprecover.texinfo +++ b/doc/lziprecover.texinfo @@ -5,8 +5,8 @@ @finalout @c %**end of header -@set UPDATED 12 November 2011 -@set VERSION 1.13-rc1 +@set UPDATED 20 November 2011 +@set VERSION 1.13-rc2 @dircategory Data Compression @direntry @@ -122,6 +122,20 @@ data as possible when decompressing a corrupt file. @itemx --decompress Decompress. +@item -D @var{range} +@itemx --range-decompress=@var{range} +Decompress only a range of bytes starting at decompressed byte position +@samp{@var{begin}} and up to byte position @w{@samp{@var{end} - 1}}. +Three formats of @var{range} are recognized, @samp{@var{begin}}, +@samp{@var{begin}-@var{end}}, and @samp{@var{begin},@var{size}}. If only +@var{begin} is specified, @var{end} is taken as the end of the file. The +produced bytes are sent to standard output unless the @samp{--output} +option is used. 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 verified. This operation is more +efficient in multimember files because it only decompresses the members +containing the desired data. + @item -f @itemx --force Force overwrite of output files. @@ -130,6 +144,11 @@ Force overwrite of output files. @itemx --keep Keep (don't delete) input files during decompression. +@item -l +@itemx --list +Print total file sizes and ratios. The values produced are correct even +for multimember files. + @item -m @itemx --merge Try to produce a correct file merging the good parts of two or more @@ -195,6 +214,24 @@ trailing garbage (if any). @end table +@sp 1 +Numbers given as arguments to options may be followed by a multiplier +and an optional @samp{B} for "byte". + +Table of SI and binary prefixes (unit multipliers): + +@multitable {Prefix} {kilobyte (10^3 = 1000)} {|} {Prefix} {kibibyte (2^10 = 1024)} +@item Prefix @tab Value @tab | @tab Prefix @tab Value +@item k @tab kilobyte (10^3 = 1000) @tab | @tab Ki @tab kibibyte (2^10 = 1024) +@item M @tab megabyte (10^6) @tab | @tab Mi @tab mebibyte (2^20) +@item G @tab gigabyte (10^9) @tab | @tab Gi @tab gibibyte (2^30) +@item T @tab terabyte (10^12) @tab | @tab Ti @tab tebibyte (2^40) +@item P @tab petabyte (10^15) @tab | @tab Pi @tab pebibyte (2^50) +@item E @tab exabyte (10^18) @tab | @tab Ei @tab exbibyte (2^60) +@item Z @tab zettabyte (10^21) @tab | @tab Zi @tab zebibyte (2^70) +@item Y @tab yottabyte (10^24) @tab | @tab Yi @tab yobibyte (2^80) +@end multitable + @node File Format @chapter File Format diff --git a/lzip.h b/lzip.h index 440ce71..1c67344 100644 --- a/lzip.h +++ b/lzip.h @@ -60,6 +60,7 @@ enum { min_dictionary_size = 1 << min_dictionary_bits, max_dictionary_bits = 29, max_dictionary_size = 1 << max_dictionary_bits, + min_member_size = 36, literal_context_bits = 3, pos_state_bits = 2, pos_states = 1 << pos_state_bits, @@ -128,6 +129,15 @@ public: if( longest_name == 0 ) longest_name = stdin_name_len; } + Pretty_print( const std::string & filename, const int v ) + : stdin_name( "(stdin)" ), verbosity_( v ), first_post( false ) + { + const unsigned int stdin_name_len = std::strlen( stdin_name ); + longest_name = ( ( filename == "-" ) ? stdin_name_len : filename.size() ); + if( longest_name == 0 ) longest_name = stdin_name_len; + set_name( filename ); + } + void set_name( const std::string & filename ) { if( filename.size() && filename != "-" ) name_ = filename; @@ -288,12 +298,35 @@ struct Error #endif +class Block + { + long long pos_, size_; // pos + size <= LLONG_MAX + +public: + Block( const long long p, const long long s ) throw() + : pos_( p ), size_( s ) {} + + long long pos() const throw() { return pos_; } + long long size() const throw() { return size_; } + long long end() const throw() { return pos_ + size_; } + + void pos( const long long p ) throw() { pos_ = p; } + void size( const long long s ) throw() { size_ = s; } + + bool overlaps( const Block & b ) const throw() + { return ( pos_ < b.end() && b.pos_ < end() ); } + void shift( Block & b ) throw() { ++size_; ++b.pos_; --b.size_; } + }; + + // defined in decoder.cc int readblock( const int fd, uint8_t * const buf, const int size ) throw(); int writeblock( const int fd, const uint8_t * const buf, const int size ) throw(); // defined in main.cc extern int verbosity; +const char * format_num( long long num, long long limit = LLONG_MAX, + const int set_prefix = 0 ) throw(); int open_instream( const std::string & name, struct stat * const in_statsp, const bool to_stdout, const bool reg_only = false ) throw(); int open_outstream_rw( const std::string & output_filename, @@ -314,6 +347,13 @@ bool verify_single_member( const int fd, const long long file_size ); int merge_files( const std::vector< std::string > & filenames, const std::string & output_filename, const bool force ); +// defined in range_dec.cc +int list_file( const std::string & input_filename ); +int range_decompress( const std::string & input_filename, + const std::string & default_output_filename, + const std::string & range_string, + const bool to_stdout, const bool force ); + // defined in repair.cc int repair_file( const std::string & input_filename, const std::string & output_filename, const bool force ); diff --git a/main.cc b/main.cc index 2abbc0e..dec1ec5 100644 --- a/main.cc +++ b/main.cc @@ -71,8 +71,8 @@ struct { const char * from; const char * to; } const known_extensions[] = { { ".tlz", ".tar" }, { 0, 0 } }; -enum Mode { m_none, m_decompress, m_generate, m_merge, m_recover, m_repair, - m_split, m_test, m_update }; +enum Mode { m_none, m_decompress, m_generate, m_list, m_merge, m_range, + m_recover, m_repair, m_split, m_test, m_update }; std::string output_filename; int outfd = -1; @@ -87,22 +87,26 @@ void show_help() throw() std::printf( "%s - Data recovery tool and decompressor for lzipped files.\n", Program_name ); std::printf( "\nUsage: %s [options] [files]\n", invocation_name ); std::printf( "\nOptions:\n" - " -h, --help display this help and exit\n" - " -V, --version output version information and exit\n" - " -c, --stdout send decompressed output to standard output\n" - " -d, --decompress decompress\n" - " -f, --force overwrite existing output files\n" -// " -g, --generate-recover-file generate a recover file\n" - " -k, --keep keep (don't delete) input files\n" - " -m, --merge correct errors in file using several copies\n" - " -o, --output= place the output into \n" - " -q, --quiet suppress all messages\n" -// " -r, --recover correct errors in file using a recover file\n" - " -R, --repair try to repair a small error in file\n" - " -s, --split split a multimember file in single-member files\n" - " -t, --test test compressed file integrity\n" -// " -u, --update convert file from version 0 to version 1\n" - " -v, --verbose be verbose (a 2nd -v gives more)\n" + " -h, --help display this help and exit\n" + " -V, --version output version information and exit\n" + " -c, --stdout send decompressed output to standard output\n" + " -d, --decompress decompress\n" + " -D, --range-decompress= decompress only a range of bytes (N-M)\n" + " -f, --force overwrite existing output files\n" +// " -g, --generate-recover-file generate a recover file\n" + " -k, --keep keep (don't delete) input files\n" + " -l, --list print total file sizes and ratios\n" + " -m, --merge correct errors in file using several copies\n" + " -o, --output= place the output into \n" + " -q, --quiet suppress all messages\n" +// " -r, --recover correct errors in file using a recover file\n" + " -R, --repair try to repair a small error in file\n" + " -s, --split split multimember file in single-member files\n" + " -t, --test test compressed file integrity\n" +// " -u, --update convert file from version 0 to version 1\n" + " -v, --verbose be verbose (a 2nd -v gives more)\n" + "Numbers may be followed by a multiplier: k = kB = 10^3 = 1000,\n" + "Ki = KiB = 2^10 = 1024, M = 10^6, Mi = 2^20, G = 10^9, Gi = 2^30, etc...\n" "\nReport bugs to lzip-bug@nongnu.org\n" "Lziprecover home page: http://www.nongnu.org/lzip/lziprecover.html\n" ); } @@ -118,20 +122,13 @@ void show_version() throw() } -const char * format_num( long long num ) throw() +void one_file( const int argind, const int arguments ) throw() { - const char * const prefix[8] = - { "Ki", "Mi", "Gi", "Ti", "Pi", "Ei", "Zi", "Yi" }; - enum { buf_size = 16, factor = 1024 }; - static char buf[buf_size]; - const char *p = ""; - bool exact = ( num % factor == 0 ); - - for( int i = 0; i < 8 && ( llabs( num ) > 9999 || - ( exact && llabs( num ) >= factor ) ); ++i ) - { num /= factor; if( num % factor != 0 ) exact = false; p = prefix[i]; } - snprintf( buf, buf_size, "%lld %s", num, p ); - return buf; + if( argind + 1 != arguments ) + { + show_error( "You must specify exactly 1 file.", 0, true ); + std::exit( 1 ); + } } @@ -345,7 +342,7 @@ int decompress( const int infd, const Pretty_print & pp, const bool testing ) if( verbosity >= 2 ) std::fprintf( stderr, "version %d, dictionary size %7sB. ", header.version(), - format_num( header.dictionary_size() ) ); + format_num( header.dictionary_size(), 9999, -1 ) ); } LZ_decoder decoder( header, rdec, outfd ); @@ -403,6 +400,30 @@ void set_signals() throw() int verbosity = 0; +const char * format_num( long long num, long long limit, + const int set_prefix ) throw() + { + const char * const si_prefix[8] = + { "k", "M", "G", "T", "P", "E", "Z", "Y" }; + const char * const binary_prefix[8] = + { "Ki", "Mi", "Gi", "Ti", "Pi", "Ei", "Zi", "Yi" }; + static bool si = true; + static char buf[32]; + + if( set_prefix ) si = ( set_prefix > 0 ); + const int factor = ( si ? 1000 : 1024 ); + const char * const * prefix = ( si ? si_prefix : binary_prefix ); + const char * p = ""; + bool exact = ( num % factor == 0 ); + + for( int i = 0; i < 8 && ( llabs( num ) > limit || + ( exact && llabs( num ) >= factor ) ); ++i ) + { num /= factor; if( num % factor != 0 ) exact = false; p = prefix[i]; } + snprintf( buf, sizeof buf, "%lld %s", num, p ); + return buf; + } + + int open_instream( const std::string & name, struct stat * const in_statsp, const bool to_stdout, const bool reg_only ) throw() { @@ -491,25 +512,28 @@ int main( const int argc, const char * const argv[] ) bool to_stdout = false; std::string input_filename; std::string default_output_filename; + std::string range_string; std::vector< std::string > filenames; invocation_name = argv[0]; const Arg_parser::Option options[] = { - { 'c', "stdout", Arg_parser::no }, - { 'd', "decompress", Arg_parser::no }, - { 'f', "force", Arg_parser::no }, - { 'h', "help", Arg_parser::no }, - { 'k', "keep", Arg_parser::no }, - { 'm', "merge", Arg_parser::no }, - { 'o', "output", Arg_parser::yes }, - { 'q', "quiet", Arg_parser::no }, - { 'R', "repair", Arg_parser::no }, - { 's', "split", Arg_parser::no }, - { 't', "test", Arg_parser::no }, - { 'v', "verbose", Arg_parser::no }, - { 'V', "version", Arg_parser::no }, - { 0 , 0, Arg_parser::no } }; + { 'c', "stdout", Arg_parser::no }, + { 'd', "decompress", Arg_parser::no }, + { 'D', "range-decompress", Arg_parser::yes }, + { 'f', "force", Arg_parser::no }, + { 'h', "help", Arg_parser::no }, + { 'k', "keep", Arg_parser::no }, + { 'l', "list", Arg_parser::no }, + { 'm', "merge", Arg_parser::no }, + { 'o', "output", Arg_parser::yes }, + { 'q', "quiet", Arg_parser::no }, + { 'R', "repair", Arg_parser::no }, + { 's', "split", Arg_parser::no }, + { 't', "test", Arg_parser::no }, + { 'v', "verbose", Arg_parser::no }, + { 'V', "version", Arg_parser::no }, + { 0 , 0, Arg_parser::no } }; const Arg_parser parser( argc, argv, options ); if( parser.error().size() ) // bad option @@ -520,15 +544,19 @@ int main( const int argc, const char * const argv[] ) { const int code = parser.code( argind ); if( !code ) break; // no more options + const std::string & arg = parser.argument( argind ).c_str(); switch( code ) { case 'c': to_stdout = true; break; case 'd': set_mode( program_mode, m_decompress ); break; + case 'D': set_mode( program_mode, m_range ); + range_string = arg; break; case 'f': force = true; break; case 'h': show_help(); return 0; case 'k': keep_input_files = true; break; + case 'l': set_mode( program_mode, m_list ); break; case 'm': set_mode( program_mode, m_merge ); break; - case 'o': default_output_filename = parser.argument( argind ); break; + case 'o': default_output_filename = arg; break; case 'q': verbosity = -1; break; case 'R': set_mode( program_mode, m_repair ); break; case 's': set_mode( program_mode, m_split ); break; @@ -552,9 +580,10 @@ int main( const int argc, const char * const argv[] ) case m_update: case m_none: internal_error( "invalid operation" ); break; case m_decompress: break; + case m_list: + one_file( argind, parser.arguments() ); + return list_file( parser.argument( argind ) ); case m_merge: - { - std::vector< std::string > filenames; for( ; argind < parser.arguments(); ++argind ) filenames.push_back( parser.argument( argind ) ); if( filenames.size() < 2 ) @@ -562,21 +591,19 @@ int main( const int argc, const char * const argv[] ) if( !default_output_filename.size() ) default_output_filename = insert_fixed( filenames[0] ); return merge_files( filenames, default_output_filename, force ); - } break; + case m_range: + one_file( argind, parser.arguments() ); + return range_decompress( parser.argument( argind ), + default_output_filename, range_string, + to_stdout, force ); case m_repair: - { - if( argind + 1 != parser.arguments() ) - { show_error( "You must specify exactly 1 file.", 0, true ); return 1; } + one_file( argind, parser.arguments() ); if( !default_output_filename.size() ) default_output_filename = insert_fixed( parser.argument( argind ) ); return repair_file( parser.argument( argind ), default_output_filename, force ); - } break; case m_split: - { - if( argind + 1 != parser.arguments() ) - { show_error( "You must specify exactly 1 file.", 0, true ); return 1; } + one_file( argind, parser.arguments() ); return split_file( parser.argument( argind ), default_output_filename, force ); - } break; case m_test: break; } diff --git a/merge.cc b/merge.cc index 298d90b..4278fc0 100644 --- a/merge.cc +++ b/merge.cc @@ -34,24 +34,6 @@ namespace { -class Block - { - long long pos_, size_; // pos + size <= LLONG_MAX - -public: - Block( const long long p, const long long s ) throw() - : pos_( p ), size_( s ) {} - - long long pos() const throw() { return pos_; } - long long size() const throw() { return size_; } - long long end() const throw() { return pos_ + size_; } - - void pos( const long long p ) throw() { pos_ = p; } - void size( const long long s ) throw() { size_ = s; } - void shift( Block & b ) throw() { ++size_; ++b.pos_; --b.size_; } - }; - - bool copy_and_diff_file( const std::vector< int > & infd_vector, const int outfd, std::vector< Block > & block_vector ) { @@ -166,7 +148,7 @@ int open_input_files( const std::vector< std::string > & filenames, if( i == 0 ) { isize = tmp; - if( isize < 36 ) { show_error( "Input file is too short." ); return 2; } + if( isize < min_member_size ) { show_error( "Input file is too short." ); return 2; } } else if( isize != tmp ) { show_error( "Sizes of input files are different." ); return 1; } @@ -251,8 +233,7 @@ bool try_decompress( const int fd, const long long file_size, header.dictionary_size() <= max_dictionary_size ) { LZ_decoder decoder( header, rdec, -1 ); - std::vector< std::string > dummy_filenames; - Pretty_print dummy( dummy_filenames, -1 ); + Pretty_print dummy( "", -1 ); if( decoder.decode_member( dummy ) == 0 && rdec.member_position() == file_size ) return true; @@ -336,7 +317,7 @@ int merge_files( const std::vector< std::string > & filenames, if( !copy_and_diff_file( infd_vector, outfd, block_vector ) ) cleanup_and_fail( output_filename, outfd, 1 ); - if( !block_vector.size() ) + if( block_vector.size() == 0 ) { show_error( "Input files are identical. Recovery is not possible." ); cleanup_and_fail( output_filename, outfd, 2 ); } diff --git a/range_dec.cc b/range_dec.cc new file mode 100644 index 0000000..650d81a --- /dev/null +++ b/range_dec.cc @@ -0,0 +1,359 @@ +/* Lziprecover - Data recovery tool for lzipped files + Copyright (C) 2009, 2010, 2011 Antonio Diaz Diaz. + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . +*/ + +#define _FILE_OFFSET_BITS 64 + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "lzip.h" +#include "decoder.h" + + +namespace { + +class Member + { + Block dblock_, mblock_; // data block, member block + +public: + Member( const long long dp, const long long ds, + const long long mp, const long long ms ) + : dblock_( dp, ds ), mblock_( mp, ms ) {} + + const Block & dblock() const throw() { return dblock_; } + Block & dblock() throw() { return dblock_; } + const Block & mblock() const throw() { return mblock_; } + Block & mblock() throw() { return mblock_; } + }; + + +int seek_read( const int fd, uint8_t * const buf, const int size, + const long long pos ) throw() + { + if( lseek( fd, pos, SEEK_SET ) == pos ) + return readblock( fd, buf, size ); + return 0; + } + + +class Member_index + { + std::vector< Member > member_vector; + +public: + Member_index( const int infd, const long long isize ) + { + long long pos = isize; // always points to a header or EOF + File_header header; + File_trailer trailer; + while( pos >= min_member_size ) + { + if( seek_read( infd, trailer.data, File_trailer::size(), + pos - File_trailer::size() ) != File_trailer::size() ) + { show_error( "Read error", errno ); std::exit( 1 ); } + const long long member_size = trailer.member_size(); + if( member_size < min_member_size || pos < member_size ) break; + if( seek_read( infd, header.data, File_header::size, + pos - member_size ) != File_header::size ) + { show_error( "Read error", errno ); std::exit( 1 ); } + if( !header.verify_magic() || !header.verify_version() ) break; + pos -= member_size; + member_vector.push_back( Member( 0, trailer.data_size(), + pos, member_size ) ); + } + if( pos != 0 || member_vector.size() == 0 ) + { + show_error( "Member size in input file trailer is corrupt." ); + std::exit( 1 ); + } + std::reverse( member_vector.begin(), member_vector.end() ); + for( unsigned int i = 0; i < member_vector.size() - 1; ++i ) + member_vector[i+1].dblock().pos( member_vector[i].dblock().end() ); + } + + long long data_end() const throw() + { if( member_vector.size() ) return member_vector.back().dblock().end(); + else return 0; } + + const Member & member( const int i ) const throw() + { return member_vector[i]; } + const Block & dblock( const int i ) const throw() + { return member_vector[i].dblock(); } + const Block & mblock( const int i ) const throw() + { return member_vector[i].mblock(); } + int members() const throw() { return (int)member_vector.size(); } + }; + + +// Returns the number of chars read, or 0 if error. +// +int parse_long_long( const char * const ptr, long long & value ) throw() + { + char * tail; + int c = 0; + errno = 0; + value = strtoll( ptr, &tail, 0 ); + if( tail == ptr || errno ) return 0; + c = tail - ptr; + + if( ptr[c] ) + { + const int factor = ( ptr[c+1] == 'i' ) ? 1024 : 1000; + int exponent = 0; + switch( ptr[c] ) + { + case 'Y': exponent = 8; break; + case 'Z': exponent = 7; break; + case 'E': exponent = 6; break; + case 'P': exponent = 5; break; + case 'T': exponent = 4; break; + case 'G': exponent = 3; break; + case 'M': exponent = 2; break; + case 'K': if( factor == 1024 ) exponent = 1; break; + case 'k': if( factor == 1000 ) exponent = 1; break; + } + if( exponent > 0 ) + { + ++c; + if( ptr[c] == 'i' ) { ++c; if( value ) format_num( 0, 0, -1 ); } + if( ptr[c] == 'B' ) ++c; + for( int i = 0; i < exponent; ++i ) + { + if( LLONG_MAX / factor >= llabs( value ) ) value *= factor; + else return 0; + } + } + } + return c; + } + + +// Recognized formats: - , +// +void parse_range( const char * const ptr, Block & range ) throw() + { + long long value = 0; + int c = parse_long_long( ptr, value ); // pos + if( c && value >= 0 && value < LLONG_MAX && + ( ptr[c] == 0 || ptr[c] == ',' || ptr[c] == '-' ) ) + { + range.pos( value ); + if( ptr[c] == 0 ) { range.size( LLONG_MAX - value ); return; } + const bool issize = ( ptr[c] == ',' ); + c = parse_long_long( ptr + c + 1, value ); // size + if( c && value > 0 && ( issize || value > range.pos() ) ) + { + if( !issize ) value -= range.pos(); + if( LLONG_MAX - range.pos() >= value ) { range.size( value ); return; } + } + } + show_error( "Bad decompression range.", 0, true ); + std::exit( 1 ); + } + + +bool safe_seek( const int fd, const long long pos ) throw() + { + if( lseek( fd, pos, SEEK_SET ) == pos ) return true; + show_error( "Seek error", errno ); return false; + } + + +int decompress_member( const int infd, const int outfd, + const Pretty_print & pp, const Member & member, + const long long outskip, const long long outend ) + { + int retval = 0; + + try { + Range_decoder rdec( infd ); + File_header header; + int size; + for( size = 0; size < File_header::size && !rdec.finished(); ++size ) + header.data[size] = rdec.get_byte(); + if( rdec.finished() ) // End Of File + { pp( "Error reading member header" ); retval = 1; } + if( !header.verify_magic() ) + { pp( "Bad magic number (file not in lzip format)" ); retval = 2; } + if( !header.verify_version() ) + { + if( verbosity >= 0 ) + { pp(); + std::fprintf( stderr, "Version %d member format not supported.\n", + header.version() ); } + retval = 2; + } + if( header.dictionary_size() < min_dictionary_size || + header.dictionary_size() > max_dictionary_size ) + { pp( "Invalid dictionary size in member header" ); retval = 2; } + + if( pp.verbosity() >= 2 ) + { + pp(); + std::fprintf( stderr, "version %d, dictionary size %7sB. ", + header.version(), + format_num( header.dictionary_size(), 9999, -1 ) ); + } + LZ_decoder decoder( header, rdec, outfd, outskip, outend ); + + const int result = decoder.decode_member( pp ); + if( result != 0 ) + { + if( verbosity >= 0 && result <= 2 ) + { + pp(); + if( result == 2 ) + std::fprintf( stderr, "File ends unexpectedly at pos %lld\n", + member.mblock().pos() + rdec.member_position() ); + else + std::fprintf( stderr, "Decoder error at pos %lld\n", + member.mblock().pos() + rdec.member_position() ); + } + retval = 2; + } + if( pp.verbosity() >= 2 ) std::fprintf( stderr, "done\n" ); + } + catch( std::bad_alloc ) + { + pp( "Not enough memory. Find a machine with more memory" ); + retval = 1; + } + catch( Error e ) { pp(); show_error( e.msg, errno ); retval = 1; } + return retval; + } + +} // end namespace + + +int list_file( const std::string & input_filename ) + { + struct stat in_stats; + const int infd = open_instream( input_filename, &in_stats, true, true ); + if( infd < 0 ) return 1; + const long long isize = lseek( infd, 0, SEEK_END ); + if( isize < 0 ) + { show_error( "Input file is not seekable", errno ); return 1; } + if( isize < min_member_size ) + { show_error( "Input file is too short." ); return 2; } + + Member_index member_index( infd, isize ); + + if( verbosity >= 0 ) + { + if( verbosity >= 1 ) + { + std::printf( "Total members in file = %d.\n", member_index.members() ); + for( int i = 0; i < member_index.members(); ++i ) + { + const Block & db = member_index.dblock( i ); + const Block & mb = member_index.mblock( i ); + std::printf( "Member %3d data pos %9lld data size %7lld " + "member pos %9lld member size %7lld.\n", i, + db.pos(), db.size(), mb.pos(), mb.size() ); + } + std::printf( "\n" ); + } + + const long long data_size = member_index.data_end(); + if( data_size > 0 && isize > 0 ) + std::printf( "%6.3f:1, %6.3f bits/byte, %5.2f%% saved.\n", + (double)data_size / isize, + ( 8.0 * isize ) / data_size, + 100.0 * ( 1.0 - ( (double)isize / data_size ) ) ); + std::printf( "decompressed size %9lld, compressed size %8lld.\n", + data_size, isize ); + } + return 0; + } + + +int range_decompress( const std::string & input_filename, + const std::string & output_filename, + const std::string & range_string, + const bool to_stdout, const bool force ) + { + Block range( 0, 0 ); + parse_range( range_string.c_str(), range ); + struct stat in_stats; + const int infd = open_instream( input_filename, &in_stats, true, true ); + if( infd < 0 ) return 1; + const long long isize = lseek( infd, 0, SEEK_END ); + if( isize < 0 ) + { show_error( "Input file is not seekable", errno ); return 1; } + if( isize < min_member_size ) + { show_error( "Input file is too short." ); return 2; } + + Member_index member_index( infd, isize ); + if( range.end() > member_index.data_end() ) + range.size( std::max( 0LL, member_index.data_end() - range.pos() ) ); + if( range.size() <= 0 ) + { if( verbosity >= 1 ) show_error( "Nothing to do." ); return 0; } + + if( verbosity >= 1 ) + { + if( verbosity >= 2 ) + std::fprintf( stderr, "Decompressed file size = %sB\n", + format_num( member_index.data_end() ) ); + std::fprintf( stderr, "Decompressing range %sB", format_num( range.pos() ) ); + std::fprintf( stderr, " to %sB ", format_num( range.pos() + range.size() ) ); + std::fprintf( stderr, "(%sBytes)\n", format_num( range.size() ) ); + } + + int outfd = -1; + if( to_stdout || !output_filename.size() ) + outfd = STDOUT_FILENO; + else + { outfd = open_outstream_rw( output_filename, force ); + if( outfd < 0 ) return 1; } + Pretty_print pp( input_filename, 0 ); + int retval = 0; + for( int i = 0; i < member_index.members(); ++i ) + { + const Block & db = member_index.dblock( i ); + if( range.overlaps( db ) ) + { + if( verbosity >= 3 ) + std::fprintf( stderr, "Decompressing member %3d\n", i ); + const long long outskip = std::max( 0LL, range.pos() - db.pos() ); + const long long outend = std::min( db.end(), range.end() - db.pos() ); + if( !safe_seek( infd, member_index.mblock( i ).pos() ) ) + { retval = 1; break; } + retval = decompress_member( infd, outfd, pp, member_index.member( i ), + outskip, outend ); + if( retval ) cleanup_and_fail( output_filename, outfd, retval ); + pp.reset(); + } + } + if( close( outfd ) != 0 ) + { + show_error( "Error closing output file", errno ); + cleanup_and_fail( output_filename, -1, 1 ); + } + if( verbosity >= 2 ) + std::fprintf( stderr, "Byte range decompressed successfully.\n" ); + return retval; + } diff --git a/repair.cc b/repair.cc index 8a92e59..38ff932 100644 --- a/repair.cc +++ b/repair.cc @@ -24,6 +24,7 @@ #include #include #include +#include #include #include "lzip.h" @@ -38,7 +39,7 @@ int repair_file( const std::string & input_filename, const long long isize = lseek( infd, 0, SEEK_END ); if( isize < 0 ) { show_error( "Input file is not seekable", errno ); return 1; } - if( isize < 36 ) + if( isize < min_member_size ) { show_error( "Input file is too short." ); return 2; } if( !verify_single_member( infd, isize ) ) return 2; if( lseek( infd, 0, SEEK_SET ) < 0 ) diff --git a/split.cc b/split.cc index d6538f8..81e4ba8 100644 --- a/split.cc +++ b/split.cc @@ -24,6 +24,7 @@ #include #include #include +#include #include #include "lzip.h" diff --git a/testsuite/check.sh b/testsuite/check.sh index 5885ba2..5b8244c 100755 --- a/testsuite/check.sh +++ b/testsuite/check.sh @@ -21,12 +21,14 @@ if [ -d tmp ] ; then rm -rf tmp ; fi mkdir tmp cd "${objdir}"/tmp -ln "${testdir}"/test.txt in || framework_failure -ln "${testdir}"/test_v1.lz in.lz || framework_failure -cat in.lz in.lz in.lz > in3.lz || framework_failure -for i in 1 2 3 4 5 ; do - ln "${testdir}"/test_bad${i}.lz bad${i}.lz -done +in="${testdir}"/test.txt +in_lz="${testdir}"/test_v1.lz +inD="${testdir}"/test921-1921.txt +bad1_lz="${testdir}"/test_bad1.lz +bad2_lz="${testdir}"/test_bad2.lz +bad3_lz="${testdir}"/test_bad3.lz +bad4_lz="${testdir}"/test_bad4.lz +bad5_lz="${testdir}"/test_bad5.lz fail=0 # Description of test files for lziprecover: @@ -38,58 +40,65 @@ fail=0 printf "testing lziprecover-%s..." "$2" -"${LZIPRECOVER}" -m -o copy.lz bad1.lz bad2.lz bad1.lz -q +"${LZIPRECOVER}" -D 921-1921 -o copy ${in_lz} || fail=1 +cmp ${inD} copy || fail=1 +printf . +"${LZIPRECOVER}" -D 921,1000 ${in_lz} > copy || fail=1 +cmp ${inD} copy || fail=1 +printf . + +"${LZIPRECOVER}" -m -o copy.lz ${bad1_lz} ${bad2_lz} ${bad1_lz} -q if [ $? != 1 ] ; then fail=1 ; printf - ; else printf . ; fi -"${LZIPRECOVER}" -m -o copy.lz bad1.lz bad2.lz || fail=1 +"${LZIPRECOVER}" -m -o copy.lz ${bad1_lz} ${bad2_lz} || fail=1 "${LZIPRECOVER}" -df copy.lz || fail=1 -cmp in copy || fail=1 +cmp ${in} copy || fail=1 printf . -"${LZIPRECOVER}" -m -o copy.lz bad2.lz bad1.lz || fail=1 +"${LZIPRECOVER}" -m -o copy.lz ${bad2_lz} ${bad1_lz} || fail=1 "${LZIPRECOVER}" -df copy.lz || fail=1 -cmp in copy || fail=1 +cmp ${in} copy || fail=1 printf . -for i in 1 2 ; do - for j in 3 4 5 ; do - "${LZIPRECOVER}" -m -o copy.lz bad${i}.lz bad${j}.lz || fail=1 +for i in ${bad1_lz} ${bad2_lz} ; do + for j in ${bad3_lz} ${bad4_lz} ${bad5_lz} ; do + "${LZIPRECOVER}" -m -o copy.lz ${i} ${j} || fail=1 "${LZIPRECOVER}" -df copy.lz || fail=1 - cmp in copy || fail=1 + cmp ${in} copy || fail=1 printf . - "${LZIPRECOVER}" -m -o copy.lz bad${j}.lz bad${i}.lz || fail=1 + "${LZIPRECOVER}" -m -o copy.lz ${j} ${i} || fail=1 "${LZIPRECOVER}" -df copy.lz || fail=1 - cmp in copy || fail=1 + cmp ${in} copy || fail=1 printf . done done -"${LZIPRECOVER}" -m -o copy.lz bad3.lz bad4.lz bad5.lz || fail=1 +"${LZIPRECOVER}" -m -o copy.lz ${bad3_lz} ${bad4_lz} ${bad5_lz} || fail=1 "${LZIPRECOVER}" -df copy.lz || fail=1 -cmp in copy || fail=1 +cmp ${in} copy || fail=1 printf . -"${LZIPRECOVER}" -m -o copy.lz bad4.lz bad5.lz bad3.lz || fail=1 +"${LZIPRECOVER}" -m -o copy.lz ${bad4_lz} ${bad5_lz} ${bad3_lz} || fail=1 "${LZIPRECOVER}" -df copy.lz || fail=1 -cmp in copy || fail=1 +cmp ${in} copy || fail=1 printf . -"${LZIPRECOVER}" -m -o copy.lz bad5.lz bad3.lz bad4.lz || fail=1 +"${LZIPRECOVER}" -m -o copy.lz ${bad5_lz} ${bad3_lz} ${bad4_lz} || fail=1 "${LZIPRECOVER}" -df copy.lz || fail=1 -cmp in copy || fail=1 +cmp ${in} copy || fail=1 printf . -"${LZIPRECOVER}" -R in.lz || fail=1 +"${LZIPRECOVER}" -R ${in_lz} || fail=1 printf . -"${LZIPRECOVER}" -R -o copy.lz bad2.lz -q +"${LZIPRECOVER}" -R -o copy.lz ${bad2_lz} -q if [ $? != 2 ] ; then fail=1 ; printf - ; else printf . ; fi -"${LZIPRECOVER}" -R -o copy.lz bad1.lz || fail=1 +"${LZIPRECOVER}" -R -o copy.lz ${bad1_lz} || fail=1 "${LZIPRECOVER}" -df copy.lz || fail=1 -cmp in copy || fail=1 +cmp ${in} copy || fail=1 printf . -cat in3.lz > out || framework_failure -printf "garbage" >> out || fail=1 -"${LZIPRECOVER}" -s out -o out.lz || fail=1 +cat ${in_lz} ${in_lz} ${in_lz} > copy || framework_failure +printf "garbage" >> copy || fail=1 +"${LZIPRECOVER}" -s -o copy.lz copy || fail=1 for i in 1 2 3 ; do - "${LZIPRECOVER}" -cd rec0000${i}out.lz > copy || fail=1 - cmp in copy || fail=1 + "${LZIPRECOVER}" -cd rec0000${i}copy.lz > copy || fail=1 + cmp ${in} copy || fail=1 printf . done diff --git a/testsuite/test921-1921.txt b/testsuite/test921-1921.txt new file mode 100644 index 0000000..70d4860 --- /dev/null +++ b/testsuite/test921-1921.txt @@ -0,0 +1,17 @@ + When we speak of free software, we are referring to freedom, not +price. Our General Public Licenses are designed to make sure that you +have the freedom to distribute copies of free software (and charge for +them if you wish), that you receive source code or can get it if you +want it, that you can change the software or use pieces of it in new +free programs, and that you know you can do these things. + + To protect your rights, we need to prevent others from denying you +these rights or asking you to surrender the rights. Therefore, you have +certain responsibilities if you distribute copies of the software, or if +you modify it: responsibilities to respect the freedom of others. + + For example, if you distribute copies of such a program, whether +gratis or for a fee, you must pass on to the recipients the same +freedoms that you received. You must make sure that they, too, receive +or can get the source code. And you must show them these terms so they +know their rights. -- cgit v1.2.3