summaryrefslogtreecommitdiffstats
path: root/range_dec.cc
diff options
context:
space:
mode:
authorDaniel Baumann <mail@daniel-baumann.ch>2015-11-07 10:20:52 +0000
committerDaniel Baumann <mail@daniel-baumann.ch>2015-11-07 10:20:52 +0000
commit9e1481f1341f12a41b988d6dcb63e8256d1e54a3 (patch)
tree55304e8361dd48c9465c4670d9c5fa3747050e86 /range_dec.cc
parentAdding upstream version 1.13. (diff)
downloadlziprecover-9e1481f1341f12a41b988d6dcb63e8256d1e54a3.tar.xz
lziprecover-9e1481f1341f12a41b988d6dcb63e8256d1e54a3.zip
Adding upstream version 1.14~rc1.upstream/1.14_rc1
Signed-off-by: Daniel Baumann <mail@daniel-baumann.ch>
Diffstat (limited to 'range_dec.cc')
-rw-r--r--range_dec.cc245
1 files changed, 89 insertions, 156 deletions
diff --git a/range_dec.cc b/range_dec.cc
index d8e171a..d056271 100644
--- a/range_dec.cc
+++ b/range_dec.cc
@@ -1,5 +1,5 @@
/* Lziprecover - Data recovery tool for lzipped files
- Copyright (C) 2009, 2010, 2011, 2012 Antonio Diaz Diaz.
+ Copyright (C) 2009, 2010, 2011, 2012, 2013 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
@@ -19,7 +19,6 @@
#include <algorithm>
#include <cerrno>
-#include <climits>
#include <cstdio>
#include <cstdlib>
#include <cstring>
@@ -31,83 +30,11 @@
#include "lzip.h"
#include "decoder.h"
+#include "file_index.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 { return dblock_; }
- Block & dblock() { return dblock_; }
- const Block & mblock() const { return mblock_; }
- Block & mblock() { return mblock_; }
- };
-
-
-int seek_read( const int fd, uint8_t * const buf, const int size,
- const long long pos )
- {
- 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
- { if( member_vector.size() ) return member_vector.back().dblock().end();
- else return 0; }
-
- const Member & member( const int i ) const { return member_vector[i]; }
- const Block & dblock( const int i ) const
- { return member_vector[i].dblock(); }
- const Block & mblock( const int i ) const
- { return member_vector[i].mblock(); }
- int members() const { 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 )
@@ -115,7 +42,7 @@ int parse_long_long( const char * const ptr, long long & value )
char * tail;
errno = 0;
value = strtoll( ptr, &tail, 0 );
- if( tail == ptr || errno ) return 0;
+ if( tail == ptr || errno || value < 0 ) return 0;
int c = tail - ptr;
if( ptr[c] )
@@ -141,7 +68,7 @@ int parse_long_long( const char * const ptr, long long & value )
if( ptr[c] == 'B' ) ++c;
for( int i = 0; i < exponent; ++i )
{
- if( LLONG_MAX / factor >= llabs( value ) ) value *= factor;
+ if( INT64_MAX / factor >= value ) value *= factor;
else return 0;
}
}
@@ -156,17 +83,17 @@ void parse_range( const char * const ptr, Block & range )
{
long long value = 0;
int c = parse_long_long( ptr, value ); // pos
- if( c && value >= 0 && value < LLONG_MAX &&
+ if( c && value >= 0 && value < INT64_MAX &&
( ptr[c] == 0 || ptr[c] == ',' || ptr[c] == '-' ) )
{
range.pos( value );
- if( ptr[c] == 0 ) { range.size( LLONG_MAX - value ); return; }
+ if( ptr[c] == 0 ) { range.size( INT64_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; }
+ if( INT64_MAX - range.pos() >= value ) { range.size( value ); return; }
}
}
show_error( "Bad decompression range.", 0, true );
@@ -182,132 +109,138 @@ bool safe_seek( const int fd, const long long pos )
int decompress_member( const int infd, const int outfd,
- const Pretty_print & pp, const Member & member,
- const long long outskip, const long long outend )
+ const Pretty_print & pp,
+ const unsigned long long mpos,
+ const unsigned long long outskip,
+ const unsigned 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();
+ rdec.read_data( header.data, File_header::size );
if( rdec.finished() ) // End Of File
- { pp( "Error reading member header" ); retval = 1; }
+ { pp( "Error reading member header" ); return 1; }
if( !header.verify_magic() )
- { pp( "Bad magic number (file not in lzip format)" ); retval = 2; }
+ { pp( "Bad magic number (file not in lzip format)" ); return 2; }
if( !header.verify_version() )
{
- if( verbosity >= 0 )
+ if( pp.verbosity() >= 0 )
{ pp();
std::fprintf( stderr, "Version %d member format not supported.\n",
header.version() ); }
- retval = 2;
+ return 2;
}
if( header.dictionary_size() < min_dictionary_size ||
header.dictionary_size() > max_dictionary_size )
- { pp( "Invalid dictionary size in member header" ); retval = 2; }
+ { pp( "Invalid dictionary size in member header" ); return 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 );
+ if( pp.verbosity() >= 2 ) { pp(); show_header( header ); }
+ LZ_decoder decoder( header, rdec, outfd, outskip, outend );
const int result = decoder.decode_member( pp );
if( result != 0 )
{
- if( verbosity >= 0 && result <= 2 )
+ if( pp.verbosity() >= 0 && result <= 2 )
{
pp();
if( result == 2 )
- std::fprintf( stderr, "File ends unexpectedly at pos %lld\n",
- member.mblock().pos() + rdec.member_position() );
+ std::fprintf( stderr, "File ends unexpectedly at pos %llu\n",
+ mpos + rdec.member_position() );
else
- std::fprintf( stderr, "Decoder error at pos %lld\n",
- member.mblock().pos() + rdec.member_position() );
+ std::fprintf( stderr, "Decoder error at pos %llu\n",
+ mpos + rdec.member_position() );
}
- retval = 2;
+ return 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;
+ return 1;
}
- catch( Error e ) { pp(); show_error( e.msg, errno ); retval = 1; }
- return retval;
+ catch( Error e ) { pp(); show_error( e.msg, errno ); return 1; }
+ return 0;
}
-} // end namespace
-
-int list_file( const std::string & input_filename )
+int list_file( const std::string & input_filename, const Pretty_print & pp )
{
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 );
+ File_index file_index( infd );
+ close( infd );
+ if( file_index.retval() != 0 )
+ { show_error( file_index.error().c_str() ); return file_index.retval(); }
- if( verbosity >= 0 )
+ if( pp.verbosity() >= 0 )
{
- if( verbosity >= 1 )
+ const unsigned long long data_size = file_index.data_end();
+ const unsigned long long file_size = file_index.file_end();
+ pp( 0, stdout );
+ if( data_size > 0 && file_size > 0 )
+ std::printf( "%6.3f:1, %6.3f bits/byte, %5.2f%% saved. ",
+ (double)data_size / file_size,
+ ( 8.0 * file_size ) / data_size,
+ 100.0 * ( 1.0 - ( (double)file_size / data_size ) ) );
+ std::printf( "decompressed size %9llu, compressed size %8llu.\n",
+ data_size, file_size );
+
+ if( pp.verbosity() >= 1 && file_index.members() > 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( "Total members in file = %d.\n", file_index.members() );
+ if( pp.verbosity() >= 2 )
+ for( int i = 0; i < file_index.members(); ++i )
+ {
+ const Block & db = file_index.dblock( i );
+ const Block & mb = file_index.mblock( i );
+ std::printf( "Member %3d data pos %9llu data size %7llu "
+ "member pos %9llu member size %7llu.\n", i + 1,
+ 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;
}
+} // end namespace
+
+
+int list_files( const std::vector< std::string > & filenames,
+ const int verbosity )
+ {
+ Pretty_print pp( filenames, verbosity );
+ int retval = 0;
+ for( unsigned i = 0; i < filenames.size(); ++i )
+ {
+ pp.set_name( filenames[i] );
+ const int tmp = list_file( filenames[i], pp );
+ if( tmp > retval ) retval = tmp;
+ }
+ return retval;
+ }
+
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 )
+ const std::string & range_string, const int verbosity,
+ const bool force, const bool to_stdout )
{
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() ) );
+
+ File_index file_index( infd );
+ if( file_index.retval() != 0 )
+ { show_error( file_index.error().c_str() ); return file_index.retval(); }
+
+ if( range.end() > file_index.data_end() )
+ range.size( std::max( 0LL, file_index.data_end() - range.pos() ) );
if( range.size() <= 0 )
{ if( verbosity >= 1 ) show_error( "Nothing to do." ); return 0; }
@@ -315,7 +248,7 @@ int range_decompress( const std::string & input_filename,
{
if( verbosity >= 2 )
std::fprintf( stderr, "Decompressed file size = %sB\n",
- format_num( member_index.data_end() ) );
+ format_num( file_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() ) );
@@ -329,23 +262,23 @@ int range_decompress( const std::string & input_filename,
if( outfd < 0 ) return 1; }
Pretty_print pp( input_filename, 0 );
int retval = 0;
- for( int i = 0; i < member_index.members(); ++i )
+ for( int i = 0; i < file_index.members(); ++i )
{
- const Block & db = member_index.dblock( i );
+ const Block & db = file_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 );
+ const long long mpos = file_index.mblock( i ).pos();
+ if( !safe_seek( infd, mpos ) ) { retval = 1; break; }
+ retval = decompress_member( infd, outfd, pp, mpos, outskip, outend );
if( retval ) cleanup_and_fail( output_filename, outfd, retval );
pp.reset();
}
}
+ close( infd );
if( close( outfd ) != 0 )
{
show_error( "Error closing output file", errno );