/* Plzip - Parallel compressor compatible with lzip
Copyright (C) 2009 Laszlo Ersek.
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
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 .
*/
/*
Exit status: 0 for a normal exit, 1 for environmental problems
(file not found, invalid flags, I/O errors, etc), 2 to indicate a
corrupt or invalid input file, 3 for an internal consistency error
(eg, bug) which caused plzip to panic.
*/
#define _FILE_OFFSET_BITS 64
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#if defined(__MSVCRT__)
#include
#define fchmod(x,y) 0
#define fchown(x,y,z) 0
#define strtoull std::strtoul
#define SIGHUP SIGTERM
#define S_ISSOCK(x) 0
#define S_IRGRP 0
#define S_IWGRP 0
#define S_IROTH 0
#define S_IWOTH 0
#endif
#if defined(__OS2__)
#include
#endif
#include "arg_parser.h"
#include "lzip.h"
#if CHAR_BIT != 8
#error "Environments where CHAR_BIT != 8 are not supported."
#endif
namespace {
const char * const Program_name = "Plzip";
const char * const program_name = "plzip";
const char * const program_year = "2013";
const char * invocation_name = 0;
#ifdef O_BINARY
const int o_binary = O_BINARY;
#else
const int o_binary = 0;
#endif
struct { const char * from; const char * to; } const known_extensions[] = {
{ ".lz", "" },
{ ".tlz", ".tar" },
{ 0, 0 } };
struct Lzma_options
{
int dictionary_size; // 4 KiB .. 512 MiB
int match_len_limit; // 5 .. 273
};
enum Mode { m_compress, m_decompress, m_test };
std::string output_filename;
int outfd = -1;
const mode_t usr_rw = S_IRUSR | S_IWUSR;
const mode_t all_rw = usr_rw | S_IRGRP | S_IWGRP | S_IROTH | S_IWOTH;
mode_t outfd_mode = usr_rw;
bool delete_output_on_interrupt = false;
void show_help( const long num_online )
{
std::printf( "%s - Parallel compressor compatible with lzip.\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"
" -B, --data-size= set input data block size in bytes\n"
" -c, --stdout send output to standard output\n"
" -d, --decompress decompress\n"
" -f, --force overwrite existing output files\n"
" -F, --recompress force recompression of compressed files\n"
" -k, --keep keep (don't delete) input files\n"
" -m, --match-length= set match length limit in bytes [36]\n"
" -n, --threads= set number of (de)compression threads [%ld]\n"
" -o, --output= if reading stdin, place the output into \n"
" -q, --quiet suppress all messages\n"
" -s, --dictionary-size= set dictionary size limit in bytes [8 MiB]\n"
" -t, --test test compressed file integrity\n"
" -v, --verbose be verbose (a 2nd -v gives more)\n"
" -1 .. -9 set compression level [default 6]\n"
" --fast alias for -1\n"
" --best alias for -9\n", num_online );
if( verbosity > 0 )
{
std::printf( " -D, --debug= (0-1) print debug statistics to stderr\n" );
}
std::printf( "If no file names are given, plzip compresses or decompresses\n"
"from standard input to standard output.\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"
"The bidimensional parameter space of LZMA can't be mapped to a linear\n"
"scale optimal for all files. If your files are large, very repetitive,\n"
"etc, you may need to use the --match-length and --dictionary-size\n"
"options directly to achieve optimal performance. For example, -9m64\n"
"usually compresses executables more (and faster) than -9.\n"
"\nExit status: 0 for a normal exit, 1 for environmental problems (file\n"
"not found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or\n"
"invalid input file, 3 for an internal consistency error (eg, bug) which\n"
"caused plzip to panic.\n"
"\nReport bugs to lzip-bug@nongnu.org\n"
"Plzip home page: http://www.nongnu.org/lzip/plzip.html\n" );
}
void show_version()
{
std::printf( "%s %s\n", Program_name, PROGVERSION );
std::printf( "Copyright (C) 2009 Laszlo Ersek.\n"
"Copyright (C) %s Antonio Diaz Diaz.\n", program_year );
std::printf( "Using Lzlib %s\n", LZ_version() );
std::printf( "License GPLv3+: GNU GPL version 3 or later \n"
"This is free software: you are free to change and redistribute it.\n"
"There is NO WARRANTY, to the extent permitted by law.\n" );
}
unsigned long long getnum( const char * const ptr,
const unsigned long long llimit,
const unsigned long long ulimit )
{
char * tail;
errno = 0;
unsigned long long result = strtoull( ptr, &tail, 0 );
if( tail == ptr )
{
show_error( "Bad or missing numerical argument.", 0, true );
std::exit( 1 );
}
if( !errno && tail[0] )
{
int factor = ( tail[1] == 'i' ) ? 1024 : 1000;
int exponent = 0;
bool bad_multiplier = false;
switch( tail[0] )
{
case ' ': break;
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; else bad_multiplier = true;
break;
case 'k': if( factor == 1000 ) exponent = 1; else bad_multiplier = true;
break;
default : bad_multiplier = true;
}
if( bad_multiplier )
{
show_error( "Bad multiplier in numerical argument.", 0, true );
std::exit( 1 );
}
for( int i = 0; i < exponent; ++i )
{
if( ulimit / factor >= result ) result *= factor;
else { errno = ERANGE; break; }
}
}
if( !errno && ( result < llimit || result > ulimit ) ) errno = ERANGE;
if( errno )
{
show_error( "Numerical argument out of limits." );
std::exit( 1 );
}
return result;
}
int get_dict_size( const char * const arg )
{
char * tail;
int bits = std::strtol( arg, &tail, 0 );
if( bits >= LZ_min_dictionary_bits() &&
bits <= LZ_max_dictionary_bits() && *tail == 0 )
return ( 1 << bits );
return getnum( arg, LZ_min_dictionary_size(), LZ_max_dictionary_size() );
}
int extension_index( const std::string & name )
{
for( int i = 0; known_extensions[i].from; ++i )
{
const std::string ext( known_extensions[i].from );
if( name.size() > ext.size() &&
name.compare( name.size() - ext.size(), ext.size(), ext ) == 0 )
return i;
}
return -1;
}
int open_instream( const char * const name, struct stat * const in_statsp,
const Mode program_mode, const int eindex,
const bool recompress, const bool to_stdout )
{
int infd = -1;
if( program_mode == m_compress && !recompress && eindex >= 0 )
{
if( verbosity >= 0 )
std::fprintf( stderr, "%s: Input file '%s' already has '%s' suffix.\n",
program_name, name, known_extensions[eindex].from );
}
else
{
infd = open( name, O_RDONLY | o_binary );
if( infd < 0 )
{
if( verbosity >= 0 )
std::fprintf( stderr, "%s: Can't open input file '%s': %s.\n",
program_name, name, std::strerror( errno ) );
}
else
{
const int i = fstat( infd, in_statsp );
const mode_t mode = in_statsp->st_mode;
const bool can_read = ( i == 0 &&
( S_ISBLK( mode ) || S_ISCHR( mode ) ||
S_ISFIFO( mode ) || S_ISSOCK( mode ) ) );
const bool no_ofile = to_stdout || program_mode == m_test;
if( i != 0 || ( !S_ISREG( mode ) && ( !can_read || !no_ofile ) ) )
{
if( verbosity >= 0 )
std::fprintf( stderr, "%s: Input file '%s' is not a regular file%s.\n",
program_name, name,
( can_read && !no_ofile ) ?
" and '--stdout' was not specified" : "" );
close( infd );
infd = -1;
}
}
}
return infd;
}
void set_c_outname( const std::string & name )
{
output_filename = name;
output_filename += known_extensions[0].from;
}
void set_d_outname( const std::string & name, const int i )
{
if( i >= 0 )
{
const std::string from( known_extensions[i].from );
if( name.size() > from.size() )
{
output_filename.assign( name, 0, name.size() - from.size() );
output_filename += known_extensions[i].to;
return;
}
}
output_filename = name; output_filename += ".out";
if( verbosity >= 1 )
std::fprintf( stderr, "%s: Can't guess original name for '%s' -- using '%s'.\n",
program_name, name.c_str(), output_filename.c_str() );
}
bool open_outstream( const bool force )
{
int flags = O_CREAT | O_WRONLY | o_binary;
if( force ) flags |= O_TRUNC; else flags |= O_EXCL;
outfd = open( output_filename.c_str(), flags, outfd_mode );
if( outfd < 0 && verbosity >= 0 )
{
if( errno == EEXIST )
std::fprintf( stderr, "%s: Output file '%s' already exists, skipping.\n",
program_name, output_filename.c_str() );
else
std::fprintf( stderr, "%s: Can't create output file '%s': %s.\n",
program_name, output_filename.c_str(), std::strerror( errno ) );
}
return ( outfd >= 0 );
}
bool check_tty( const int infd, const Mode program_mode )
{
if( program_mode == m_compress && outfd >= 0 && isatty( outfd ) )
{
show_error( "I won't write compressed data to a terminal.", 0, true );
return false;
}
if( ( program_mode == m_decompress || program_mode == m_test ) &&
isatty( infd ) )
{
show_error( "I won't read compressed data from a terminal.", 0, true );
return false;
}
return true;
}
// Set permissions, owner and times.
void close_and_set_permissions( const struct stat * const in_statsp )
{
bool warning = false;
if( in_statsp )
{
// fchown will in many cases return with EPERM, which can be safely ignored.
if( ( fchown( outfd, in_statsp->st_uid, in_statsp->st_gid ) != 0 &&
errno != EPERM ) ||
fchmod( outfd, in_statsp->st_mode ) != 0 ) warning = true;
}
if( close( outfd ) != 0 ) cleanup_and_fail( 1 );
outfd = -1;
delete_output_on_interrupt = false;
if( in_statsp )
{
struct utimbuf t;
t.actime = in_statsp->st_atime;
t.modtime = in_statsp->st_mtime;
if( utime( output_filename.c_str(), &t ) != 0 ) warning = true;
}
if( warning && verbosity >= 1 )
show_error( "Can't change output file attributes." );
}
extern "C" void signal_handler( int )
{
show_error( "Control-C or similar caught, quitting." );
cleanup_and_fail( 1 );
}
void set_signals()
{
std::signal( SIGHUP, signal_handler );
std::signal( SIGINT, signal_handler );
std::signal( SIGTERM, signal_handler );
}
} // end namespace
int verbosity = 0;
void Pretty_print::operator()( const char * const msg ) const
{
if( verbosity >= 0 )
{
if( first_post )
{
first_post = false;
std::fprintf( stderr, " %s: ", name_.c_str() );
for( unsigned i = 0; i < longest_name - name_.size(); ++i )
std::fprintf( stderr, " " );
if( !msg ) std::fflush( stderr );
}
if( msg ) std::fprintf( stderr, "%s.\n", msg );
}
}
void show_error( const char * const msg, const int errcode, const bool help )
{
if( verbosity >= 0 )
{
if( msg && msg[0] )
{
std::fprintf( stderr, "%s: %s", program_name, msg );
if( errcode > 0 )
std::fprintf( stderr, ": %s", std::strerror( errcode ) );
std::fprintf( stderr, "\n" );
}
if( help )
std::fprintf( stderr, "Try '%s --help' for more information.\n",
invocation_name );
}
}
void internal_error( const char * const msg )
{
if( verbosity >= 0 )
std::fprintf( stderr, "%s: internal error: %s.\n", program_name, msg );
std::exit( 3 );
}
// This can be called from any thread, main thread or sub-threads alike,
// since they all call common helper functions that call cleanup_and_fail()
// in case of an error.
//
void cleanup_and_fail( const int retval )
{
// only one thread can delete and exit
static pthread_mutex_t mutex = PTHREAD_MUTEX_INITIALIZER;
pthread_mutex_lock( &mutex ); // ignore errors to avoid loop
if( delete_output_on_interrupt )
{
delete_output_on_interrupt = false;
if( verbosity >= 0 )
std::fprintf( stderr, "%s: Deleting output file '%s', if it exists.\n",
program_name, output_filename.c_str() );
if( outfd >= 0 ) { close( outfd ); outfd = -1; }
if( std::remove( output_filename.c_str() ) != 0 && errno != ENOENT )
show_error( "WARNING: deletion of output file (apparently) failed." );
}
std::exit( retval );
}
void show_progress( const int packet_size,
const Pretty_print * const p,
const struct stat * const in_statsp )
{
static unsigned long long cfile_size = 0; // file_size / 100
static unsigned long long pos = 0;
static const Pretty_print * pp = 0;
static pthread_mutex_t mutex;
if( p ) // initialize static vars
{
if( !pp ) xinit( &mutex ); // init mutex only once
pos = 0; pp = p;
cfile_size = ( in_statsp && S_ISREG( in_statsp->st_mode ) ) ?
in_statsp->st_size / 100 : 0;
return;
}
if( pp )
{
xlock( &mutex );
pos += packet_size;
if( cfile_size > 0 )
std::fprintf( stderr, "%4llu%%", pos / cfile_size );
std::fprintf( stderr, " %.1f MB\r", pos / 1000000.0 );
pp->reset(); (*pp)(); // restore cursor position
xunlock( &mutex );
}
}
int main( const int argc, const char * const argv[] )
{
// Mapping from gzip/bzip2 style 1..9 compression modes
// to the corresponding LZMA compression modes.
const Lzma_options option_mapping[] =
{
{ 1 << 20, 5 }, // -0
{ 1 << 20, 5 }, // -1
{ 3 << 19, 6 }, // -2
{ 1 << 21, 8 }, // -3
{ 3 << 20, 12 }, // -4
{ 1 << 22, 20 }, // -5
{ 1 << 23, 36 }, // -6
{ 1 << 24, 68 }, // -7
{ 3 << 23, 132 }, // -8
{ 1 << 25, 273 } }; // -9
Lzma_options encoder_options = option_mapping[6]; // default = "-6"
std::string input_filename;
std::string default_output_filename;
std::vector< std::string > filenames;
int data_size = 0;
int debug_level = 0;
int infd = -1;
int num_workers = 0; // start this many worker threads
Mode program_mode = m_compress;
bool force = false;
bool keep_input_files = false;
bool recompress = false;
bool to_stdout = false;
invocation_name = argv[0];
if( LZ_version()[0] != LZ_version_string[0] )
internal_error( "bad library version" );
const long num_online = std::max( 1L, sysconf( _SC_NPROCESSORS_ONLN ) );
long max_workers = sysconf( _SC_THREAD_THREADS_MAX );
if( max_workers < 1 || max_workers > INT_MAX / (int)sizeof (pthread_t) )
max_workers = INT_MAX / sizeof (pthread_t);
const Arg_parser::Option options[] =
{
{ '0', "fast", Arg_parser::no },
{ '1', 0, Arg_parser::no },
{ '2', 0, Arg_parser::no },
{ '3', 0, Arg_parser::no },
{ '4', 0, Arg_parser::no },
{ '5', 0, Arg_parser::no },
{ '6', 0, Arg_parser::no },
{ '7', 0, Arg_parser::no },
{ '8', 0, Arg_parser::no },
{ '9', "best", Arg_parser::no },
{ 'b', "member-size", Arg_parser::yes },
{ 'B', "data-size", Arg_parser::yes },
{ 'c', "stdout", Arg_parser::no },
{ 'd', "decompress", Arg_parser::no },
{ 'D', "debug", Arg_parser::yes },
{ 'f', "force", Arg_parser::no },
{ 'F', "recompress", Arg_parser::no },
{ 'h', "help", Arg_parser::no },
{ 'k', "keep", Arg_parser::no },
{ 'm', "match-length", Arg_parser::yes },
{ 'n', "threads", Arg_parser::yes },
{ 'o', "output", Arg_parser::yes },
{ 'q', "quiet", Arg_parser::no },
{ 's', "dictionary-size", Arg_parser::yes },
{ 'S', "volume-size", Arg_parser::yes },
{ '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
{ show_error( parser.error().c_str(), 0, true ); return 1; }
int argind = 0;
for( ; argind < parser.arguments(); ++argind )
{
const int code = parser.code( argind );
if( !code ) break; // no more options
const char * const arg = parser.argument( argind ).c_str();
switch( code )
{
case '0':
case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
encoder_options = option_mapping[code-'0']; break;
case 'b': break;
case 'B': data_size = getnum( arg, 2 * LZ_min_dictionary_size(),
2 * LZ_max_dictionary_size() ); break;
case 'c': to_stdout = true; break;
case 'd': program_mode = m_decompress; break;
case 'D': debug_level = getnum( arg, 0, 3 ); break;
case 'f': force = true; break;
case 'F': recompress = true; break;
case 'h': show_help( num_online ); return 0;
case 'k': keep_input_files = true; break;
case 'm': encoder_options.match_len_limit =
getnum( arg, LZ_min_match_len_limit(),
LZ_max_match_len_limit() ); break;
case 'n': num_workers = getnum( arg, 1, max_workers ); break;
case 'o': default_output_filename = arg; break;
case 'q': verbosity = -1; break;
case 's': encoder_options.dictionary_size = get_dict_size( arg );
break;
case 'S': break;
case 't': program_mode = m_test; break;
case 'v': if( verbosity < 4 ) ++verbosity; break;
case 'V': show_version(); return 0;
default : internal_error( "uncaught option" );
}
} // end process options
#if defined(__MSVCRT__) || defined(__OS2__)
setmode( STDIN_FILENO, O_BINARY );
setmode( STDOUT_FILENO, O_BINARY );
#endif
if( program_mode == m_test )
outfd = -1;
if( data_size <= 0 )
data_size = 2 * std::max( 65536, encoder_options.dictionary_size );
else if( data_size < encoder_options.dictionary_size )
encoder_options.dictionary_size = std::max( data_size, LZ_min_dictionary_size() );
if( num_workers <= 0 )
num_workers = std::min( num_online, max_workers );
bool filenames_given = false;
for( ; argind < parser.arguments(); ++argind )
{
filenames.push_back( parser.argument( argind ) );
if( filenames.back() != "-" ) filenames_given = true;
}
if( filenames.empty() ) filenames.push_back("-");
if( !to_stdout && program_mode != m_test &&
( filenames_given || default_output_filename.size() ) )
set_signals();
Pretty_print pp( filenames );
int retval = 0;
for( unsigned i = 0; i < filenames.size(); ++i )
{
struct stat in_stats;
output_filename.clear();
if( !filenames[i].size() || filenames[i] == "-" )
{
input_filename.clear();
infd = STDIN_FILENO;
if( program_mode != m_test )
{
if( to_stdout || !default_output_filename.size() )
outfd = STDOUT_FILENO;
else
{
if( program_mode == m_compress )
set_c_outname( default_output_filename );
else output_filename = default_output_filename;
outfd_mode = all_rw;
if( !open_outstream( force ) )
{
if( retval < 1 ) retval = 1;
close( infd ); infd = -1;
continue;
}
}
}
}
else
{
input_filename = filenames[i];
const int eindex = extension_index( input_filename );
infd = open_instream( input_filename.c_str(), &in_stats, program_mode,
eindex, recompress, to_stdout );
if( infd < 0 ) { if( retval < 1 ) retval = 1; continue; }
if( program_mode != m_test )
{
if( to_stdout ) outfd = STDOUT_FILENO;
else
{
if( program_mode == m_compress )
set_c_outname( input_filename );
else set_d_outname( input_filename, eindex );
outfd_mode = usr_rw;
if( !open_outstream( force ) )
{
if( retval < 1 ) retval = 1;
close( infd ); infd = -1;
continue;
}
}
}
}
if( !check_tty( infd, program_mode ) ) return 1;
if( output_filename.size() && !to_stdout && program_mode != m_test )
delete_output_on_interrupt = true;
const struct stat * const in_statsp = input_filename.size() ? &in_stats : 0;
const bool infd_isreg = in_statsp && S_ISREG( in_statsp->st_mode );
pp.set_name( input_filename );
if( verbosity >= 1 ) pp();
int tmp;
if( program_mode == m_compress )
{
show_progress( 0, &pp, in_statsp ); // initialize static vars
if( verbosity >= 2 ) show_progress( 0 ); // show initial zero size
tmp = compress( data_size, encoder_options.dictionary_size,
encoder_options.match_len_limit,
num_workers, infd, outfd, pp, debug_level );
}
else
tmp = decompress( num_workers, infd, outfd, pp, debug_level,
program_mode == m_test, infd_isreg );
if( tmp > retval ) retval = tmp;
if( tmp && program_mode != m_test ) cleanup_and_fail( retval );
if( delete_output_on_interrupt )
close_and_set_permissions( in_statsp );
if( input_filename.size() )
{
close( infd ); infd = -1;
if( !keep_input_files && !to_stdout && program_mode != m_test )
std::remove( input_filename.c_str() );
}
}
if( outfd >= 0 && close( outfd ) != 0 )
{
show_error( "Can't close stdout", errno );
if( retval < 1 ) retval = 1;
}
return retval;
}