diff options
Diffstat (limited to '')
| -rw-r--r-- | main.c | 173 |
1 files changed, 108 insertions, 65 deletions
@@ -1,5 +1,5 @@ /* Clzip - LZMA lossless data compressor - Copyright (C) 2010-2021 Antonio Diaz Diaz. + Copyright (C) 2010-2022 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 @@ -18,7 +18,7 @@ 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 clzip to panic. + (e.g., bug) which caused clzip to panic. */ #define _FILE_OFFSET_BITS 64 @@ -36,9 +36,9 @@ #include <unistd.h> #include <utime.h> #include <sys/stat.h> -#if defined(__MSVCRT__) || defined(__OS2__) || defined(__DJGPP__) +#if defined __MSVCRT__ || defined __OS2__ || defined __DJGPP__ #include <io.h> -#if defined(__MSVCRT__) +#if defined __MSVCRT__ #define fchmod(x,y) 0 #define fchown(x,y,z) 0 #define strtoull strtoul @@ -51,7 +51,7 @@ #define S_IWOTH 0 #endif #endif -#if defined(__DJGPP__) +#if defined __DJGPP__ #define S_ISSOCK(x) 0 #define S_ISVTX 0 #endif @@ -72,10 +72,15 @@ #error "Environments where CHAR_BIT != 8 are not supported." #endif +#if ( defined SIZE_MAX && SIZE_MAX < UINT_MAX ) || \ + ( defined SSIZE_MAX && SSIZE_MAX < INT_MAX ) +#error "Environments where 'size_t' is narrower than 'int' are not supported." +#endif + int verbosity = 0; static const char * const program_name = "clzip"; -static const char * const program_year = "2021"; +static const char * const program_year = "2022"; static const char * invocation_name = "clzip"; /* default value */ static const struct { const char * from; const char * to; } known_extensions[] = { @@ -106,13 +111,14 @@ static void show_help( void ) "C++ compiler.\n" "\nLzip is a lossless data compressor with a user interface similar to the one\n" "of gzip or bzip2. Lzip uses a simplified form of the 'Lempel-Ziv-Markov\n" - "chain-Algorithm' (LZMA) stream format, chosen to maximize safety and\n" - "interoperability. Lzip can compress about as fast as gzip (lzip -0) or\n" - "compress most files more than bzip2 (lzip -9). Decompression speed is\n" - "intermediate between gzip and bzip2. Lzip is better than gzip and bzip2 from\n" - "a data recovery perspective. Lzip has been designed, written, and tested\n" - "with great care to replace gzip and bzip2 as the standard general-purpose\n" - "compressed format for unix-like systems.\n" + "chain-Algorithm' (LZMA) stream format and provides a 3 factor integrity\n" + "checking to maximize interoperability and optimize safety. Lzip can compress\n" + "about as fast as gzip (lzip -0) or compress most files more than bzip2\n" + "(lzip -9). Decompression speed is intermediate between gzip and bzip2.\n" + "Lzip is better than gzip and bzip2 from a data recovery perspective. Lzip\n" + "has been designed, written, and tested with great care to replace gzip and\n" + "bzip2 as the standard general-purpose compressed format for unix-like\n" + "systems.\n" "\nUsage: %s [options] [files]\n", invocation_name ); printf( "\nOptions:\n" " -h, --help display this help and exit\n" @@ -150,7 +156,7 @@ static void show_help( void ) "'tar -xf foo.tar.lz' or 'clzip -cd foo.tar.lz | tar -xf -'.\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" + "invalid input file, 3 for an internal consistency error (e.g., bug) which\n" "caused clzip to panic.\n" "\nThe ideas embodied in clzip are due to (at least) the following people:\n" "Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the\n" @@ -194,8 +200,6 @@ struct Pretty_print static void Pp_init( struct Pretty_print * const pp, const char * const filenames[], const int num_filenames ) { - unsigned stdin_name_len; - int i; pp->name = 0; pp->padded_name = 0; pp->stdin_name = "(stdin)"; @@ -203,7 +207,8 @@ static void Pp_init( struct Pretty_print * const pp, pp->first_post = false; if( verbosity <= 0 ) return; - stdin_name_len = strlen( pp->stdin_name ); + const unsigned stdin_name_len = strlen( pp->stdin_name ); + int i; for( i = 0; i < num_filenames; ++i ) { const char * const s = filenames[i]; @@ -237,16 +242,14 @@ static void Pp_reset( struct Pretty_print * const pp ) void Pp_show_msg( struct Pretty_print * const pp, const char * const msg ) { - if( verbosity >= 0 ) + if( verbosity < 0 ) return; + if( pp->first_post ) { - if( pp->first_post ) - { - pp->first_post = false; - fputs( pp->padded_name, stderr ); - if( !msg ) fflush( stderr ); - } - if( msg ) fprintf( stderr, "%s\n", msg ); + pp->first_post = false; + fputs( pp->padded_name, stderr ); + if( !msg ) fflush( stderr ); } + if( msg ) fprintf( stderr, "%s\n", msg ); } @@ -284,17 +287,53 @@ void show_header( const unsigned dictionary_size ) } -static unsigned long long getnum( const char * const ptr, +/* separate large numbers >= 100_000 in groups of 3 digits using '_' */ +static const char * format_num3( unsigned long long num ) + { + const char * const si_prefix = "kMGTPEZY"; + const char * const binary_prefix = "KMGTPEZY"; + enum { buffers = 8, bufsize = 4 * sizeof (long long) }; + static char buffer[buffers][bufsize]; /* circle of static buffers for printf */ + static int current = 0; + int i; + char * const buf = buffer[current++]; current %= buffers; + char * p = buf + bufsize - 1; /* fill the buffer backwards */ + *p = 0; /* terminator */ + if( num > 1024 ) + { + char prefix = 0; /* try binary first, then si */ + for( i = 0; i < 8 && num >= 1024 && num % 1024 == 0; ++i ) + { num /= 1024; prefix = binary_prefix[i]; } + if( prefix ) *(--p) = 'i'; + else + for( i = 0; i < 8 && num >= 1000 && num % 1000 == 0; ++i ) + { num /= 1000; prefix = si_prefix[i]; } + if( prefix ) *(--p) = prefix; + } + const bool split = num >= 100000; + + for( i = 0; ; ) + { + *(--p) = num % 10 + '0'; num /= 10; if( num == 0 ) break; + if( split && ++i >= 3 ) { i = 0; *(--p) = '_'; } + } + return p; + } + + +static unsigned long long getnum( const char * const arg, + const char * const option_name, const unsigned long long llimit, const unsigned long long ulimit ) { - unsigned long long result; char * tail; errno = 0; - result = strtoull( ptr, &tail, 0 ); - if( tail == ptr ) + unsigned long long result = strtoull( arg, &tail, 0 ); + if( tail == arg ) { - show_error( "Bad or missing numerical argument.", 0, true ); + if( verbosity >= 0 ) + fprintf( stderr, "%s: Bad or missing numerical argument in " + "option '%s'.\n", program_name, option_name ); exit( 1 ); } @@ -317,7 +356,9 @@ static unsigned long long getnum( const char * const ptr, } if( exponent <= 0 ) { - show_error( "Bad multiplier in numerical argument.", 0, true ); + if( verbosity >= 0 ) + fprintf( stderr, "%s: Bad multiplier in numerical argument of " + "option '%s'.\n", program_name, option_name ); exit( 1 ); } for( i = 0; i < exponent; ++i ) @@ -329,21 +370,24 @@ static unsigned long long getnum( const char * const ptr, if( !errno && ( result < llimit || result > ulimit ) ) errno = ERANGE; if( errno ) { - show_error( "Numerical argument out of limits.", 0, false ); + if( verbosity >= 0 ) + fprintf( stderr, "%s: Numerical argument out of limits [%s,%s] " + "in option '%s'.\n", program_name, format_num3( llimit ), + format_num3( ulimit ), option_name ); exit( 1 ); } return result; } -static int get_dict_size( const char * const arg ) +static int get_dict_size( const char * const arg, const char * const option_name ) { char * tail; const long bits = strtol( arg, &tail, 0 ); if( bits >= min_dictionary_bits && bits <= max_dictionary_bits && *tail == 0 ) return 1 << bits; - return getnum( arg, min_dictionary_size, max_dictionary_size ); + return getnum( arg, option_name, min_dictionary_size, max_dictionary_size ); } @@ -519,7 +563,7 @@ static bool check_tty_in( const char * const input_filename, const int infd, isatty( infd ) ) /* for example /dev/tty */ { show_file_error( input_filename, "I won't read compressed data from a terminal.", 0 ); - close( infd ); set_retval( retval, 1 ); + close( infd ); set_retval( retval, 2 ); if( program_mode != m_test ) cleanup_and_fail( *retval ); return false; } return true; @@ -600,7 +644,6 @@ static int compress( const unsigned long long cfile_size, struct Pretty_print * const pp, const struct stat * const in_statsp, const bool zero ) { - unsigned long long in_size = 0, out_size = 0, partial_volume_size = 0; int retval = 0; struct Poly_encoder encoder = { 0, 0, 0 }; /* polymorphic encoder */ if( verbosity >= 1 ) Pp_show_msg( pp, 0 ); @@ -633,6 +676,7 @@ static int compress( const unsigned long long cfile_size, } } + unsigned long long in_size = 0, out_size = 0, partial_volume_size = 0; while( true ) /* encode one member per iteration */ { const unsigned long long size = ( volume_size > 0 ) ? @@ -729,12 +773,9 @@ static int decompress( const unsigned long long cfile_size, const int infd, for( first_member = true; ; first_member = false ) { - int result, size; - unsigned dictionary_size; Lzip_header header; - struct LZ_decoder decoder; Rd_reset_member_position( &rdec ); - size = Rd_read_data( &rdec, header, Lh_size ); + const int size = Rd_read_data( &rdec, header, Lh_size ); if( Rd_finished( &rdec ) ) /* End Of File */ { if( first_member ) @@ -764,17 +805,18 @@ static int decompress( const unsigned long long cfile_size, const int infd, if( !Lh_verify_version( header ) ) { Pp_show_msg( pp, bad_version( Lh_version( header ) ) ); retval = 2; break; } - dictionary_size = Lh_get_dictionary_size( header ); + const unsigned dictionary_size = Lh_get_dictionary_size( header ); if( !isvalid_ds( dictionary_size ) ) { Pp_show_msg( pp, bad_dict_msg ); retval = 2; break; } if( verbosity >= 2 || ( verbosity == 1 && first_member ) ) Pp_show_msg( pp, 0 ); + struct LZ_decoder decoder; if( !LZd_init( &decoder, &rdec, dictionary_size, outfd ) ) { Pp_show_msg( pp, mem_msg ); retval = 1; break; } show_dprogress( cfile_size, partial_file_pos, &rdec, pp ); /* init */ - result = LZd_decode_member( &decoder, pp ); + const int result = LZd_decode_member( &decoder, pp ); partial_file_pos += Rd_member_position( &rdec ); LZd_free( &decoder ); if( result != 0 ) @@ -911,24 +953,16 @@ int main( const int argc, const char * const argv[] ) unsigned long long member_size = max_member_size; unsigned long long volume_size = 0; const char * default_output_filename = ""; - static struct Arg_parser parser; /* static because valgrind complains */ - static struct Pretty_print pp; /* and memory management in C sucks */ - static const char ** filenames = 0; - int num_filenames = 0; enum Mode program_mode = m_compress; - int argind = 0; - int failed_tests = 0; - int retval = 0; int i; - bool filenames_given = false; bool force = false; bool ignore_trailing = true; bool keep_input_files = false; bool loose_trailing = false; bool recompress = false; - bool stdin_used = false; bool to_stdout = false; bool zero = false; + if( argc > 0 ) invocation_name = argv[0]; enum { opt_lt = 256 }; const struct ap_Option options[] = @@ -964,19 +998,22 @@ int main( const int argc, const char * const argv[] ) { opt_lt, "loose-trailing", ap_no }, { 0, 0, ap_no } }; - if( argc > 0 ) invocation_name = argv[0]; CRC32_init(); + /* static because valgrind complains and memory management in C sucks */ + static struct Arg_parser parser; if( !ap_init( &parser, argc, argv, options, 0 ) ) { show_error( mem_msg, 0, false ); return 1; } if( ap_error( &parser ) ) /* bad option */ { show_error( ap_error( &parser ), 0, true ); return 1; } + int argind = 0; for( ; argind < ap_arguments( &parser ); ++argind ) { const int code = ap_code( &parser, argind ); - const char * const arg = ap_argument( &parser, argind ); if( !code ) break; /* no more options */ + const char * const pn = ap_parsed_name( &parser, argind ); + const char * const arg = ap_argument( &parser, argind ); switch( code ) { case '0': case '1': case '2': case '3': case '4': @@ -984,7 +1021,7 @@ int main( const int argc, const char * const argv[] ) zero = ( code == '0' ); encoder_options = option_mapping[code-'0']; break; case 'a': ignore_trailing = false; break; - case 'b': member_size = getnum( arg, 100000, max_member_size ); break; + case 'b': member_size = getnum( arg, pn, 100000, max_member_size ); break; case 'c': to_stdout = true; break; case 'd': set_mode( &program_mode, m_decompress ); break; case 'f': force = true; break; @@ -993,15 +1030,15 @@ int main( const int argc, const char * const argv[] ) case 'k': keep_input_files = true; break; case 'l': set_mode( &program_mode, m_list ); break; case 'm': encoder_options.match_len_limit = - getnum( arg, min_match_len_limit, max_match_len ); + getnum( arg, pn, min_match_len_limit, max_match_len ); zero = false; break; case 'n': break; case 'o': if( strcmp( arg, "-" ) == 0 ) to_stdout = true; else { default_output_filename = arg; } break; case 'q': verbosity = -1; break; - case 's': encoder_options.dictionary_size = get_dict_size( arg ); + case 's': encoder_options.dictionary_size = get_dict_size( arg, pn ); zero = false; break; - case 'S': volume_size = getnum( arg, 100000, max_volume_size ); break; + case 'S': volume_size = getnum( arg, pn, 100000, max_volume_size ); break; case 't': set_mode( &program_mode, m_test ); break; case 'v': if( verbosity < 4 ) ++verbosity; break; case 'V': show_version(); return 0; @@ -1010,15 +1047,17 @@ int main( const int argc, const char * const argv[] ) } } /* end process options */ -#if defined(__MSVCRT__) || defined(__OS2__) || defined(__DJGPP__) +#if defined __MSVCRT__ || defined __OS2__ || defined __DJGPP__ setmode( STDIN_FILENO, O_BINARY ); setmode( STDOUT_FILENO, O_BINARY ); #endif - num_filenames = max( 1, ap_arguments( &parser ) - argind ); + static const char ** filenames = 0; + int num_filenames = max( 1, ap_arguments( &parser ) - argind ); filenames = resize_buffer( filenames, num_filenames * sizeof filenames[0] ); filenames[0] = "-"; + bool filenames_given = false; for( i = 0; argind + i < ap_arguments( &parser ); ++i ) { filenames[i] = ap_argument( &parser, argind + i ); @@ -1052,17 +1091,18 @@ int main( const int argc, const char * const argv[] ) if( !to_stdout && program_mode != m_test && ( filenames_given || to_file ) ) set_signals( signal_handler ); + static struct Pretty_print pp; Pp_init( &pp, filenames, num_filenames ); + int failed_tests = 0; + int retval = 0; const bool one_to_one = !to_stdout && program_mode != m_test && !to_file; + bool stdin_used = false; for( i = 0; i < num_filenames; ++i ) { - unsigned long long cfile_size; const char * input_filename = ""; int infd; - int tmp; struct stat in_stats; - const struct stat * in_statsp; Pp_set_name( &pp, filenames[i] ); if( strcmp( filenames[i], "-" ) == 0 ) @@ -1104,9 +1144,12 @@ int main( const int argc, const char * const argv[] ) return 1; /* check tty only once and don't try to delete a tty */ } - in_statsp = ( input_filename[0] && one_to_one ) ? &in_stats : 0; - cfile_size = ( input_filename[0] && S_ISREG( in_stats.st_mode ) ) ? - ( in_stats.st_size + 99 ) / 100 : 0; + const struct stat * const in_statsp = + ( input_filename[0] && one_to_one ) ? &in_stats : 0; + const unsigned long long cfile_size = + ( input_filename[0] && S_ISREG( in_stats.st_mode ) ) ? + ( in_stats.st_size + 99 ) / 100 : 0; + int tmp; if( program_mode == m_compress ) tmp = compress( cfile_size, member_size, volume_size, infd, &encoder_options, &pp, in_statsp, zero ); |