diff options
Diffstat (limited to 'main.c')
| -rw-r--r-- | main.c | 1249 |
1 files changed, 1249 insertions, 0 deletions
@@ -0,0 +1,1249 @@ +/* Pdlzip - LZMA lossless data compressor + 2009-08-14 : Igor Pavlov : Public domain + Copyright (C) 2010-2024 Antonio Diaz Diaz. + + This program is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: + + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. + + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + 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. +*/ +/* + Exit status: 0 for a normal exit, 1 for environmental problems + (file not found, invalid command-line options, I/O errors, etc), 2 to + indicate a corrupt or invalid input file, 3 for an internal consistency + error (e.g., bug) which caused pdlzip to panic. +*/ + +#define _FILE_OFFSET_BITS 64 + +#include <errno.h> +#include <fcntl.h> +#include <limits.h> /* SSIZE_MAX */ +#include <signal.h> +#include <stdbool.h> +#include <stdint.h> /* SIZE_MAX */ +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <unistd.h> +#include <utime.h> +#include <sys/stat.h> +#if defined __MSVCRT__ || defined __OS2__ || defined __DJGPP__ +#include <io.h> +#if defined __MSVCRT__ +#define fchmod(x,y) 0 +#define fchown(x,y,z) 0 +#define SIGHUP SIGTERM +#define S_ISSOCK(x) 0 +#ifndef S_IRGRP +#define S_IRGRP 0 +#define S_IWGRP 0 +#define S_IROTH 0 +#define S_IWOTH 0 +#endif +#endif +#if defined __DJGPP__ +#define S_ISSOCK(x) 0 +#define S_ISVTX 0 +#endif +#endif + +#include "carg_parser.h" +#include "lzip.h" +#include "LzmaDec.h" +#include "LzmaEnc.h" + +#ifndef O_BINARY +#define O_BINARY 0 +#endif + +#if CHAR_BIT != 8 +#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 void cleanup_and_fail( const int retval ); +static void show_error( const char * const msg, const int errcode, + const bool help ); +static void show_file_error( const char * const filename, + const char * const msg, const int errcode ); +static void internal_error( const char * const msg ); + +static const char * const program_name = "pdlzip"; +static const char * const program_year = "2024"; +static const char * invocation_name = "pdlzip"; /* default value */ + +static const struct { const char * from; const char * to; } known_extensions[] = { + { ".lz", "" }, + { ".tlz", ".tar" }, + { ".lzma", "" }, + { 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 }; + +/* Variables used in signal handler context. + They are not declared volatile because the handler never returns. */ +static char * output_filename = 0; +static int outfd = -1; +static bool delete_output_on_interrupt = false; + + +static void show_help( void ) + { + printf( "Pdlzip is a permissively licensed implementation of the lzip data\n" + "compressor, intended for those who can't distribute (or even use) GPL\n" + "licensed Free Software. The name of pdlzip comes from 'public domain lzip'.\n" + "Pdlzip is written in C and is compatible with lzip 1.4 or newer.\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 to maximize interoperability. The\n" + "maximum dictionary size is 512 MiB so that any lzip file can be decompressed\n" + "on 32-bit machines. Lzip provides accurate and robust 3-factor integrity\n" + "checking. Lzip can compress about as fast as gzip (lzip -0) or compress most\n" + "files more than bzip2 (lzip -9). Decompression speed is intermediate between\n" + "gzip and bzip2. Lzip is better than gzip and bzip2 from a data recovery\n" + "perspective. Lzip has been designed, written, and tested with great care to\n" + "replace gzip and bzip2 as the standard general-purpose compressed format for\n" + "Unix-like systems.\n" + "\nPdlzip is also able to decompress legacy lzma-alone (.lzma) files.\n" + "Lzma-alone is a very bad format; it is essentially a raw LZMA stream.\n" + "If you keep any lzma-alone files, it is advisable to recompress them to\n" + "lzip format. Lziprecover can convert some lzma-alone files to lzip format\n" + "without recompressing.\n" + "\nUsage: %s [options] [files]\n", invocation_name ); + printf( "\nOptions:\n" + " -h, --help display this help and exit\n" + " -V, --version output version information and exit\n" + " -a, --trailing-error exit with error status if trailing data\n" + " -c, --stdout write to standard output, keep input files\n" + " -d, --decompress decompress, test compressed file integrity\n" + " -f, --force overwrite existing output files\n" + " -F, --recompress force re-compression of compressed files\n" + " -k, --keep keep (don't delete) input files\n" + " -m, --match-length=<bytes> set match length limit in bytes [36]\n" + " -o, --output=<file> write to <file>, keep input files\n" + " -q, --quiet suppress all messages\n" + " -s, --dictionary-size=<bytes> 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" + " -0 .. -9 set compression level [default 6]\n" + " --fast alias for -0\n" + " --best alias for -9\n" + " --loose-trailing allow trailing data seeming corrupt header\n" + "\nIf no file names are given, or if a file is '-', pdlzip compresses or\n" + "decompresses 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" + "Dictionary sizes 12 to 27 are interpreted as powers of two, meaning 2^12 to\n" + "2^27 bytes.\n" + "\nThe bidimensional parameter space of LZMA can't be mapped to a linear scale\n" + "optimal for all files. If your files are large, very repetitive, etc, you\n" + "may need to use the options --dictionary-size and --match-length directly\n" + "to achieve optimal performance. For example, -9m64 usually compresses\n" + "executables more (and faster) than -9.\n" + "\nTo extract all the files from archive 'foo.tar.lz', use the commands\n" + "'tar -xf foo.tar.lz' or 'pdlzip -cd foo.tar.lz | tar -xf -'.\n" + "\nExit status: 0 for a normal exit, 1 for environmental problems\n" + "(file not found, invalid command-line options, I/O errors, etc), 2 to\n" + "indicate a corrupt or invalid input file, 3 for an internal consistency\n" + "error (e.g., bug) which caused pdlzip to panic.\n" + "\nPdlzip includes public domain compression/decompression code from the LZMA\n" + "SDK (Software Development Kit) written by Igor Pavlov.\n" + "\nReport bugs to lzip-bug@nongnu.org\n" + "Pdlzip home page: http://www.nongnu.org/lzip/pdlzip.html\n" ); + } + + +static void show_version( void ) + { + printf( "%s %s\n", program_name, PROGVERSION ); + printf( "Copyright (C) %s Antonio Diaz Diaz.\n", program_year ); + printf( "Public Domain 2009 Igor Pavlov.\n" + "License 2-clause BSD.\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" ); + } + + +/* assure at least a minimum size for buffer 'buf' */ +static void * resize_buffer( void * buf, const unsigned min_size ) + { + if( buf ) buf = realloc( buf, min_size ); + else buf = malloc( min_size ); + if( !buf ) { show_error( mem_msg, 0, false ); cleanup_and_fail( 1 ); } + return buf; + } + + +struct Pretty_print + { + const char * name; + char * padded_name; + const char * stdin_name; + unsigned longest_name; + bool first_post; + }; + +static void Pp_init( struct Pretty_print * const pp, + const char * const filenames[], const int num_filenames ) + { + pp->name = 0; + pp->padded_name = 0; + pp->stdin_name = "(stdin)"; + pp->longest_name = 0; + pp->first_post = false; + + if( verbosity <= 0 ) return; + const unsigned stdin_name_len = strlen( pp->stdin_name ); + int i; + for( i = 0; i < num_filenames; ++i ) + { + const char * const s = filenames[i]; + const unsigned len = (strcmp( s, "-" ) == 0) ? stdin_name_len : strlen( s ); + if( pp->longest_name < len ) pp->longest_name = len; + } + if( pp->longest_name == 0 ) pp->longest_name = stdin_name_len; + } + +static void Pp_set_name( struct Pretty_print * const pp, + const char * const filename ) + { + unsigned name_len, padded_name_len, i = 0; + + if( filename && filename[0] && strcmp( filename, "-" ) != 0 ) + pp->name = filename; + else pp->name = pp->stdin_name; + name_len = strlen( pp->name ); + padded_name_len = max( name_len, pp->longest_name ) + 4; + pp->padded_name = resize_buffer( pp->padded_name, padded_name_len + 1 ); + while( i < 2 ) pp->padded_name[i++] = ' '; + while( i < name_len + 2 ) { pp->padded_name[i] = pp->name[i-2]; ++i; } + pp->padded_name[i++] = ':'; + while( i < padded_name_len ) pp->padded_name[i++] = ' '; + pp->padded_name[i] = 0; + pp->first_post = true; + } + +static void Pp_reset( struct Pretty_print * const pp ) + { if( pp->name && pp->name[0] ) pp->first_post = true; } + +static void Pp_show_msg( struct Pretty_print * const pp, const char * const msg ) + { + if( verbosity < 0 ) return; + if( pp->first_post ) + { + pp->first_post = false; + fputs( pp->padded_name, stderr ); + if( !msg ) fflush( stderr ); + } + if( msg ) fprintf( stderr, "%s\n", msg ); + } + + +static void show_header( const unsigned dictionary_size ) + { + enum { factor = 1024, n = 3 }; + const char * const prefix[n] = { "Ki", "Mi", "Gi" }; + const char * p = ""; + const char * np = " "; + unsigned num = dictionary_size; + bool exact = ( num % factor == 0 ); + + int i; for( i = 0; i < n && ( num > 9999 || ( exact && num >= factor ) ); ++i ) + { num /= factor; if( num % factor != 0 ) exact = false; + p = prefix[i]; np = ""; } + fprintf( stderr, "dict %s%4u %sB, ", np, num, p ); + } + + +/* separate numbers of 5 or more digits in groups of 3 digits using '_' */ +static const char * format_num3( unsigned long long num ) + { + enum { buffers = 8, bufsize = 4 * sizeof num, n = 10 }; + const char * const si_prefix = "kMGTPEZYRQ"; + const char * const binary_prefix = "KMGTPEZYRQ"; + 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 < n && num != 0 && num % 1024 == 0; ++i ) + { num /= 1024; prefix = binary_prefix[i]; } + if( prefix ) *(--p) = 'i'; + else + for( i = 0; i < n && num != 0 && num % 1000 == 0; ++i ) + { num /= 1000; prefix = si_prefix[i]; } + if( prefix ) *(--p) = prefix; + } + const bool split = num >= 10000; + + for( i = 0; ; ) + { + *(--p) = num % 10 + '0'; num /= 10; if( num == 0 ) break; + if( split && ++i >= 3 ) { i = 0; *(--p) = '_'; } + } + return p; + } + + +void show_option_error( const char * const arg, const char * const msg, + const char * const option_name ) + { + if( verbosity >= 0 ) + fprintf( stderr, "%s: '%s': %s option '%s'.\n", + program_name, arg, msg, option_name ); + } + + +/* Recognized formats: <num>k, <num>Ki, <num>[MGTPEZYRQ][i] */ +static unsigned long getnum( const char * const arg, + const char * const option_name, + const unsigned long llimit, + const unsigned long ulimit ) + { + char * tail; + errno = 0; + unsigned long result = strtoul( arg, &tail, 0 ); + if( tail == arg ) + { show_option_error( arg, "Bad or missing numerical argument in", + option_name ); exit( 1 ); } + + if( !errno && tail[0] ) + { + const unsigned factor = ( tail[1] == 'i' ) ? 1024 : 1000; + int exponent = 0; /* 0 = bad multiplier */ + int i; + switch( tail[0] ) + { + case 'Q': exponent = 10; break; + case 'R': exponent = 9; 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; break; + case 'k': if( factor == 1000 ) exponent = 1; break; + } + if( exponent <= 0 ) + { show_option_error( arg, "Bad multiplier in numerical argument of", + option_name ); exit( 1 ); } + for( 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 ) + { + if( verbosity >= 0 ) + fprintf( stderr, "%s: '%s': Value out of limits [%s,%s] in " + "option '%s'.\n", program_name, arg, format_num3( llimit ), + format_num3( ulimit ), option_name ); + exit( 1 ); + } + return result; + } + + +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_c && *tail == 0 ) + return 1 << bits; + return getnum( arg, option_name, min_dictionary_size, max_dictionary_size_c ); + } + + +static void set_mode( enum Mode * const program_modep, const enum Mode new_mode ) + { + if( *program_modep != m_compress && *program_modep != new_mode ) + { + show_error( "Only one operation can be specified.", 0, true ); + exit( 1 ); + } + *program_modep = new_mode; + } + + +static int extension_index( const char * const name ) + { + int eindex; + for( eindex = 0; known_extensions[eindex].from; ++eindex ) + { + const char * const ext = known_extensions[eindex].from; + const unsigned name_len = strlen( name ); + const unsigned ext_len = strlen( ext ); + if( name_len > ext_len && + strncmp( name + name_len - ext_len, ext, ext_len ) == 0 ) + return eindex; + } + return -1; + } + + +static void set_c_outname( const char * const name, const bool filenames_given, + const bool force_ext ) + { + /* zupdate < 1.9 depends on lzip adding the extension '.lz' to name when + reading from standard input. */ + output_filename = resize_buffer( output_filename, strlen( name ) + + strlen( known_extensions[0].from ) + 1 ); + strcpy( output_filename, name ); + if( force_ext || + ( !filenames_given && extension_index( output_filename ) < 0 ) ) + strcat( output_filename, known_extensions[0].from ); + } + + +static void set_d_outname( const char * const name, const int eindex ) + { + const unsigned name_len = strlen( name ); + if( eindex >= 0 ) + { + const char * const from = known_extensions[eindex].from; + const unsigned from_len = strlen( from ); + if( name_len > from_len ) + { + output_filename = resize_buffer( output_filename, name_len + + strlen( known_extensions[eindex].to ) + 1 ); + strcpy( output_filename, name ); + strcpy( output_filename + name_len - from_len, known_extensions[eindex].to ); + return; + } + } + output_filename = resize_buffer( output_filename, name_len + 4 + 1 ); + strcpy( output_filename, name ); + strcat( output_filename, ".out" ); + if( verbosity >= 1 ) + fprintf( stderr, "%s: %s: Can't guess original name -- using '%s'\n", + program_name, name, output_filename ); + } + + +static int open_instream( const char * const name, struct stat * const in_statsp, + const enum Mode program_mode, const int eindex, + const bool one_to_one, const bool recompress ) + { + if( program_mode == m_compress && !recompress && eindex >= 0 ) + { + if( verbosity >= 0 ) + fprintf( stderr, "%s: %s: Input file already has '%s' suffix.\n", + program_name, name, known_extensions[eindex].from ); + return -1; + } + int infd = open( name, O_RDONLY | O_BINARY ); + if( infd < 0 ) + show_file_error( name, "Can't open input file", 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 ) ) ); + if( i != 0 || ( !S_ISREG( mode ) && ( !can_read || one_to_one ) ) ) + { + if( verbosity >= 0 ) + fprintf( stderr, "%s: %s: Input file is not a regular file%s.\n", + program_name, name, ( can_read && one_to_one ) ? + ",\n and neither '-c' nor '-o' were specified" : "" ); + close( infd ); + infd = -1; + } + } + return infd; + } + + +static bool open_outstream( const bool force, const bool protect ) + { + const mode_t usr_rw = S_IRUSR | S_IWUSR; + const mode_t all_rw = usr_rw | S_IRGRP | S_IWGRP | S_IROTH | S_IWOTH; + const mode_t outfd_mode = protect ? usr_rw : all_rw; + int flags = O_CREAT | O_WRONLY | O_BINARY; + if( force ) flags |= O_TRUNC; else flags |= O_EXCL; + + outfd = open( output_filename, flags, outfd_mode ); + if( outfd >= 0 ) delete_output_on_interrupt = true; + else if( errno == EEXIST ) + show_file_error( output_filename, + "Output file already exists, skipping.", 0 ); + else + show_file_error( output_filename, "Can't create output file", errno ); + return outfd >= 0; + } + + +static void set_signals( void (*action)(int) ) + { + signal( SIGHUP, action ); + signal( SIGINT, action ); + signal( SIGTERM, action ); + } + + +static void cleanup_and_fail( const int retval ) + { + set_signals( SIG_IGN ); /* ignore signals */ + if( delete_output_on_interrupt ) + { + delete_output_on_interrupt = false; + show_file_error( output_filename, "Deleting output file, if it exists.", 0 ); + if( outfd >= 0 ) { close( outfd ); outfd = -1; } + if( remove( output_filename ) != 0 && errno != ENOENT ) + show_error( "warning: deletion of output file failed", errno, false ); + } + exit( retval ); + } + + +static void signal_handler( int sig ) + { + if( sig ) {} /* keep compiler happy */ + show_error( "Control-C or similar caught, quitting.", 0, false ); + cleanup_and_fail( 1 ); + } + + +static bool check_tty_in( const char * const input_filename, const int infd, + const enum Mode program_mode, int * const retval ) + { + if( ( program_mode == m_decompress || program_mode == m_test ) && + 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, 2 ); + if( program_mode != m_test ) cleanup_and_fail( *retval ); + return false; } + return true; + } + +static bool check_tty_out( const enum Mode program_mode ) + { + if( program_mode == m_compress && isatty( outfd ) ) + { show_file_error( output_filename[0] ? + output_filename : "(stdout)", + "I won't write compressed data to a terminal.", 0 ); + return false; } + return true; + } + + +/* Set permissions, owner, and times. */ +static void close_and_set_permissions( const struct stat * const in_statsp ) + { + bool warning = false; + if( in_statsp ) + { + const mode_t mode = in_statsp->st_mode; + /* fchown in many cases returns with EPERM, which can be safely ignored. */ + if( fchown( outfd, in_statsp->st_uid, in_statsp->st_gid ) == 0 ) + { if( fchmod( outfd, mode ) != 0 ) warning = true; } + else + if( errno != EPERM || + fchmod( outfd, mode & ~( S_ISUID | S_ISGID | S_ISVTX ) ) != 0 ) + warning = true; + } + if( close( outfd ) != 0 ) + { show_file_error( output_filename, "Error closing output file", errno ); + 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, &t ) != 0 ) warning = true; + } + if( warning && verbosity >= 1 ) + show_file_error( output_filename, + "warning: can't change output file attributes", errno ); + } + + +static int compress( const int infd, const struct Lzma_options * const + encoder_options, struct Pretty_print * const pp ) + { + int retval = 0; + CLzmaEncHandle encoder = 0; + Lzip_header header; + Lh_set_magic( header ); + + if( verbosity >= 1 ) Pp_show_msg( pp, 0 ); + if( Lh_set_dictionary_size( header, encoder_options->dictionary_size ) && + encoder_options->match_len_limit >= min_match_len_limit && + encoder_options->match_len_limit <= max_match_len ) + encoder = LzmaEnc_Init( Lh_get_dictionary_size( header ), + encoder_options->match_len_limit, infd, outfd ); + else internal_error( "invalid argument to encoder." ); + + if( !encoder ) + { + Pp_show_msg( pp, "Not enough memory. Try a smaller dictionary size." ); + return 1; + } + + if( writeblock( outfd, header, Lh_size ) != Lh_size ) + { show_error( "Can't write output file", errno, false ); retval = 1; } + else + if( LzmaEnc_Encode( encoder ) != 0 ) + { Pp_show_msg( pp, "Encoder error." ); retval = 1; } + LzmaEnc_Free( encoder ); + return retval; + } + + +static void show_results( const long long data_size, + const long long member_size, const unsigned crc, + const unsigned dictionary_size, const bool lzip_mode ) + { + if( verbosity >= 2 ) + { + if( verbosity >= 4 ) show_header( dictionary_size ); + if( data_size == 0 || member_size == 0 ) + fputs( "no data compressed. ", stderr ); + else + fprintf( stderr, "%6.3f:1, %5.2f%% ratio, %5.2f%% saved. ", + (double)data_size / member_size, + ( 100.0 * member_size ) / data_size, + 100.0 - ( ( 100.0 * member_size ) / data_size ) ); + if( verbosity >= 4 && lzip_mode ) fprintf( stderr, "CRC %08X, ", crc ); + if( verbosity >= 3 ) + fprintf( stderr, "%9llu out, %8llu in. ", data_size, member_size ); + if( !lzip_mode ) fputs( "lzma-alone, ", stderr ); + } + } + + +#define IN_BUF_SIZE (1 << 16) +#define OUT_BUF_SIZE (1 << 16) + +static bool read_inbuf( const int infd, uint8_t inBuf[], + int * const inPos, int * const inSize ) + { + int rest; + if( *inPos >= *inSize ) *inSize = 0; + else if( *inPos > 0 ) + { + memmove( inBuf, inBuf + *inPos, *inSize - *inPos ); + *inSize -= *inPos; + } + *inPos = 0; + rest = IN_BUF_SIZE - *inSize; + if( rest > 0 ) + { + const int rd = readblock( infd, inBuf + *inSize, rest ); + if( rd < rest && errno ) + { show_error( "Read error", errno, false ); return false; } + *inSize += rd; + } + return true; + } + + +/* 5 bytes of LZMA properties and 8 bytes of uncompressed size */ +enum { lzma_header_size = LZMA_PROPS_SIZE + 8 }; + +static int lzma_decode( uint64_t unpackSize, CLzmaDec *decoder, const int infd, + uint8_t inBuf[], int * const inPos, + int * const inSize, const unsigned dictionary_size, + const bool testing ) + { + unsigned long long member_size = lzma_header_size, data_size = 0; + uint8_t outBuf[OUT_BUF_SIZE]; + int outPos = 0; + const bool thereIsSize = (unpackSize != (uint64_t)-1); + + for (;;) + { + uint32_t inProcessed; + uint32_t outProcessed = OUT_BUF_SIZE - outPos; + ELzmaFinishMode finishMode = LZMA_FINISH_ANY; + ELzmaStatus status; + + if( *inPos >= *inSize && !read_inbuf( infd, inBuf, inPos, inSize ) ) + return 1; + inProcessed = *inSize - *inPos; + if (thereIsSize && outProcessed > unpackSize) + { + outProcessed = unpackSize; + finishMode = LZMA_FINISH_END; + } + + if( !LzmaDec_DecodeToBuf( decoder, outBuf + outPos, &outProcessed, + inBuf + *inPos, &inProcessed, finishMode, &status ) ) + { show_error( "Data error.", 0, false ); return 2; } + *inPos += inProcessed; + member_size += inProcessed; + outPos += outProcessed; + unpackSize -= outProcessed; + + if( outfd >= 0 && writeblock( outfd, outBuf, outPos ) != outPos ) + { show_error( "Can't write output file", errno, false ); return 1; } + + data_size += outPos; + outPos = 0; + + if( ( inProcessed == 0 && outProcessed == 0 ) || + ( thereIsSize && unpackSize == 0 ) ) + { + if( ( thereIsSize && unpackSize != 0 ) || + ( !thereIsSize && status != LZMA_STATUS_FINISHED_WITH_MARK ) ) + { show_error( "Data error.", 0, false ); return 2; } + show_results( data_size, member_size, 0, dictionary_size, false ); + if( verbosity >= 1 ) + fputs( testing ? "(apparently) ok\n" : "(apparently) done\n", stderr ); + return 0; + } + } + } + + +static int lzip_decode( CLzmaDec *decoder, const int infd, + struct Pretty_print * const pp, uint8_t inBuf[], + int * const inPos, int * const inSize, + const unsigned dictionary_size ) + { + unsigned long long member_size = Lh_size, data_size = 0; + uint8_t outBuf[OUT_BUF_SIZE]; + int outPos = 0; + uint32_t crc = 0xFFFFFFFFU; + + for (;;) + { + uint32_t inProcessed; + uint32_t outProcessed = OUT_BUF_SIZE - outPos; + ELzmaFinishMode finishMode = LZMA_FINISH_ANY; + ELzmaStatus status; + + if( *inPos >= *inSize && !read_inbuf( infd, inBuf, inPos, inSize ) ) + return 1; + if( *inPos >= *inSize ) + { Pp_show_msg( pp, "Unexpected EOF." ); return 2; } + inProcessed = *inSize - *inPos; + + if( !LzmaDec_DecodeToBuf( decoder, outBuf + outPos, &outProcessed, + inBuf + *inPos, &inProcessed, finishMode, &status ) ) + { Pp_show_msg( pp, "Data error." ); return 2; } + *inPos += inProcessed; + member_size += inProcessed; + outPos += outProcessed; + + if( outfd >= 0 && writeblock( outfd, outBuf, outPos ) != outPos ) + { show_error( "Can't write output file", errno, false ); return 1; } + + CRC32_update_buf( &crc, outBuf, outPos ); + data_size += outPos; + outPos = 0; + + if (inProcessed == 0 && outProcessed == 0) + { + Lzip_trailer trailer; + int i; + unsigned td_crc; + unsigned long long td_size, tm_size; + bool error = false; + + if( status != LZMA_STATUS_FINISHED_WITH_MARK ) + { Pp_show_msg( pp, "Data error." ); return 2; } + if( *inSize - *inPos < Lt_size && + !read_inbuf( infd, inBuf, inPos, inSize ) ) return 1; + if( *inSize - *inPos < Lt_size ) + { + error = true; + if( verbosity >= 0 ) + { + Pp_show_msg( pp, 0 ); + fprintf( stderr, "Trailer truncated at trailer position %d;" + " some checks may fail.\n", *inSize - *inPos ); + } + } + for( i = 0; i < Lt_size && *inPos < *inSize; ++i ) + trailer[i] = inBuf[(*inPos)++]; + member_size += i; + while( i < Lt_size ) trailer[i++] = 0; + crc ^= 0xFFFFFFFFU; + td_crc = Lt_get_data_crc( trailer ); + if( td_crc != crc ) + { + error = true; + if( verbosity >= 0 ) + { + Pp_show_msg( pp, 0 ); + fprintf( stderr, "CRC mismatch; stored %08X, computed %08X\n", + td_crc, crc ); + } + } + td_size = Lt_get_data_size( trailer ); + if( td_size != data_size ) + { + error = true; + if( verbosity >= 0 ) + { + Pp_show_msg( pp, 0 ); + fprintf( stderr, "Data size mismatch; stored %llu (0x%llX), computed %llu (0x%llX)\n", + td_size, td_size, data_size, data_size ); + } + } + tm_size = Lt_get_member_size( trailer ); + if( tm_size != member_size ) + { + error = true; + if( verbosity >= 0 ) + { + Pp_show_msg( pp, 0 ); + fprintf( stderr, "Member size mismatch; stored %llu (0x%llX), computed %llu (0x%llX)\n", + tm_size, tm_size, member_size, member_size ); + } + } + if( error ) return 2; + show_results( data_size, member_size, td_crc, dictionary_size, true ); + return 0; + } + } + } + + +static int decompress( const int infd, struct Pretty_print * const pp, + const bool ignore_trailing, const bool loose_trailing, + const bool testing ) + { + uint64_t unpackSize = 0; + CLzmaDec decoder; + uint8_t inBuf[IN_BUF_SIZE]; + int inPos = 0, inSize = 0; + int retval = 0; + bool lzip_mode = true; + bool first_member; + uint8_t raw_props[lzma_header_size]; + + for( first_member = true; ; first_member = false ) + { + int i; + unsigned dictionary_size = 0; /* keep gcc 3.3.6 quiet */ + Lzip_header header; + if( inSize - inPos < lzma_header_size && + !read_inbuf( infd, inBuf, &inPos, &inSize ) ) return 1; + const int size = inSize - inPos; + for( i = 0; i < size && i < Lh_size; ++i ) + raw_props[i] = header[i] = inBuf[inPos++]; + if( size <= Lh_size ) /* End Of File */ + { + if( first_member ) + { show_file_error( pp->name, "File ends unexpectedly at member header.", 0 ); + retval = 2; } + else if( Lh_check_prefix( header, size ) ) + { Pp_show_msg( pp, "Truncated header in multimember file." ); + retval = 2; } + else if( size > 0 && !ignore_trailing ) + { Pp_show_msg( pp, trailing_msg ); retval = 2; } + break; + } + if( !Lh_check_magic( header ) ) + { + if( !first_member ) + { + if( !loose_trailing && Lh_check_corrupt( header ) ) + { Pp_show_msg( pp, "Corrupt header in multimember file." ); + retval = 2; } + else if( !ignore_trailing ) + { Pp_show_msg( pp, trailing_msg ); retval = 2; } + break; + } + if( inSize - inPos >= lzma_header_size - Lh_size ) /* try lzma-alone */ + { + for( i = Lh_size; i < lzma_header_size; ++i ) + raw_props[i] = inBuf[inPos++]; + if( ( raw_props[12] == 0 || raw_props[12] == 0xFF ) && + raw_props[12] == raw_props[11] && raw_props[0] < (9 * 5 * 5) ) + { + lzip_mode = false; + dictionary_size = 0; + for( i = 4; i >= 1; --i ) + { dictionary_size <<= 8; dictionary_size += raw_props[i]; } + for( i = 7; i >= 0; --i ) + { unpackSize <<= 8; unpackSize += raw_props[LZMA_PROPS_SIZE+i]; } + } + } + if( lzip_mode ) + { + show_file_error( pp->name, + "Bad magic number (file not in lzip format).", 0 ); + retval = 2; break; + } + } + if( lzip_mode ) + { + if( !Lh_check_version( header ) ) + { + if( verbosity >= 0 ) + { Pp_show_msg( pp, 0 ); + fprintf( stderr, "Version %d member format not supported.\n", + Lh_version( header ) ); } + retval = 2; break; + } + dictionary_size = Lh_get_dictionary_size( header ); + if( !isvalid_ds( dictionary_size ) ) + { Pp_show_msg( pp, "Invalid dictionary size in member header." ); + retval = 2; break; } + + raw_props[0] = 93; /* (45 * 2) + (9 * 0) + 3 */ + int ds = dictionary_size; + for( i = 1; i <= 4; ++i ) { raw_props[i] = ds & 0xFF; ds >>= 8; } + } + + if( verbosity >= 2 || ( verbosity == 1 && first_member ) ) + Pp_show_msg( pp, 0 ); + + if( !LzmaDec_Init( &decoder, raw_props ) ) + { Pp_show_msg( pp, mem_msg ); return 1; } + if( lzip_mode ) + retval = lzip_decode( &decoder, infd, pp, inBuf, &inPos, &inSize, + dictionary_size ); + else + retval = lzma_decode( unpackSize, &decoder, infd, inBuf, &inPos, + &inSize, dictionary_size, testing ); + LzmaDec_Free(&decoder); + if( retval != 0 || !lzip_mode ) break; + if( verbosity >= 2 ) + { fputs( testing ? "ok\n" : "done\n", stderr ); Pp_reset( pp ); } + } + if( lzip_mode && verbosity == 1 && retval == 0 ) + fputs( testing ? "ok\n" : "done\n", stderr ); + return retval; + } + + +CRC32 crc32; + + +/* Return the number of bytes really read. + If (value returned < size) and (errno == 0), means EOF was reached. +*/ +int readblock( const int fd, uint8_t * const buf, const int size ) + { + int sz = 0; + errno = 0; + while( sz < size ) + { + const int n = read( fd, buf + sz, size - sz ); + if( n > 0 ) sz += n; + else if( n == 0 ) break; /* EOF */ + else if( errno != EINTR ) break; + errno = 0; + } + return sz; + } + + +/* Return the number of bytes really written. + If (value returned < size), it is always an error. +*/ +int writeblock( const int fd, const uint8_t * const buf, const int size ) + { + int sz = 0; + errno = 0; + while( sz < size ) + { + const int n = write( fd, buf + sz, size - sz ); + if( n > 0 ) sz += n; + else if( n < 0 && errno != EINTR ) break; + errno = 0; + } + return sz; + } + + +static void show_error( const char * const msg, const int errcode, + const bool help ) + { + if( verbosity < 0 ) return; + if( msg && msg[0] ) + fprintf( stderr, "%s: %s%s%s\n", program_name, msg, + ( errcode > 0 ) ? ": " : "", + ( errcode > 0 ) ? strerror( errcode ) : "" ); + if( help ) + fprintf( stderr, "Try '%s --help' for more information.\n", + invocation_name ); + } + + +static void show_file_error( const char * const filename, + const char * const msg, const int errcode ) + { + if( verbosity >= 0 ) + fprintf( stderr, "%s: %s: %s%s%s\n", program_name, filename, msg, + ( errcode > 0 ) ? ": " : "", + ( errcode > 0 ) ? strerror( errcode ) : "" ); + } + + +static void internal_error( const char * const msg ) + { + if( verbosity >= 0 ) + fprintf( stderr, "%s: internal error: %s\n", program_name, msg ); + exit( 3 ); + } + + +int main( const int argc, const char * const argv[] ) + { + /* Mapping from gzip/bzip2 style 0..9 compression levels to the + corresponding LZMA compression parameters. */ + const struct Lzma_options option_mapping[] = + { + { 1 << 16, 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 */ + struct Lzma_options encoder_options = option_mapping[6]; /* default = "-6" */ + const char * default_output_filename = ""; + enum Mode program_mode = m_compress; + int i; + bool force = false; + bool ignore_trailing = true; + bool keep_input_files = false; + bool loose_trailing = false; + bool recompress = false; + bool to_stdout = false; + if( argc > 0 ) invocation_name = argv[0]; + + enum { opt_lt = 256 }; + const struct ap_Option options[] = + { + { '0', "fast", ap_no }, + { '1', 0, ap_no }, + { '2', 0, ap_no }, + { '3', 0, ap_no }, + { '4', 0, ap_no }, + { '5', 0, ap_no }, + { '6', 0, ap_no }, + { '7', 0, ap_no }, + { '8', 0, ap_no }, + { '9', "best", ap_no }, + { 'a', "trailing-error", ap_no }, + { 'b', "member-size", ap_yes }, + { 'c', "stdout", ap_no }, + { 'd', "decompress", ap_no }, + { 'f', "force", ap_no }, + { 'F', "recompress", ap_no }, + { 'h', "help", ap_no }, + { 'k', "keep", ap_no }, + { 'm', "match-length", ap_yes }, + { 'n', "threads", ap_yes }, + { 'o', "output", ap_yes }, + { 'q', "quiet", ap_no }, + { 's', "dictionary-size", ap_yes }, + { 'S', "volume-size", ap_yes }, + { 't', "test", ap_no }, + { 'v', "verbose", ap_no }, + { 'V', "version", ap_no }, + { opt_lt, "loose-trailing", ap_no }, + { 0, 0, ap_no } }; + + 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 ); + 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': + case '5': case '6': case '7': case '8': case '9': + encoder_options = option_mapping[code-'0']; break; + case 'a': ignore_trailing = false; break; + case 'b': break; + case 'c': to_stdout = true; break; + case 'd': set_mode( &program_mode, m_decompress ); break; + case 'f': force = true; break; + case 'F': recompress = true; break; + case 'h': show_help(); return 0; + case 'k': keep_input_files = true; break; + case 'm': encoder_options.match_len_limit = + getnum( arg, pn, min_match_len_limit, max_match_len ); 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, pn ); + break; + case 'S': break; + case 't': set_mode( &program_mode, m_test ); break; + case 'v': if( verbosity < 4 ) ++verbosity; break; + case 'V': show_version(); return 0; + case opt_lt: loose_trailing = true; break; + default: internal_error( "uncaught option." ); + } + } /* end process options */ + +#if defined __MSVCRT__ || defined __OS2__ || defined __DJGPP__ + setmode( STDIN_FILENO, O_BINARY ); + setmode( STDOUT_FILENO, O_BINARY ); +#endif + + 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 ); + if( strcmp( filenames[i], "-" ) != 0 ) filenames_given = true; + } + + if( program_mode == m_test ) to_stdout = false; /* apply overrides */ + if( program_mode == m_test || to_stdout ) default_output_filename = ""; + + output_filename = resize_buffer( output_filename, 1 ); + output_filename[0] = 0; + if( to_stdout && program_mode != m_test ) /* check tty only once */ + { outfd = STDOUT_FILENO; if( !check_tty_out( program_mode ) ) return 1; } + else outfd = -1; + + const bool to_file = !to_stdout && program_mode != m_test && + default_output_filename[0]; + 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; + struct stat in_stats; + for( i = 0; i < num_filenames; ++i ) + { + const char * input_filename = ""; + int infd; + + Pp_set_name( &pp, filenames[i] ); + if( strcmp( filenames[i], "-" ) == 0 ) + { + if( stdin_used ) continue; else stdin_used = true; + infd = STDIN_FILENO; + if( !check_tty_in( pp.name, infd, program_mode, &retval ) ) continue; + if( one_to_one ) { outfd = STDOUT_FILENO; output_filename[0] = 0; } + } + else + { + const int eindex = extension_index( input_filename = filenames[i] ); + infd = open_instream( input_filename, &in_stats, program_mode, + eindex, one_to_one, recompress ); + if( infd < 0 ) { set_retval( &retval, 1 ); continue; } + if( !check_tty_in( pp.name, infd, program_mode, &retval ) ) continue; + if( one_to_one ) /* open outfd after checking infd */ + { + if( program_mode == m_compress ) + set_c_outname( input_filename, true, true ); + else set_d_outname( input_filename, eindex ); + if( !open_outstream( force, true ) ) + { close( infd ); set_retval( &retval, 1 ); continue; } + } + } + + if( one_to_one && !check_tty_out( program_mode ) ) + { set_retval( &retval, 1 ); return retval; } /* don't delete a tty */ + + if( to_file && outfd < 0 ) /* open outfd after checking infd */ + { + if( program_mode == m_compress ) set_c_outname( default_output_filename, + filenames_given, false ); + else + { output_filename = resize_buffer( output_filename, + strlen( default_output_filename ) + 1 ); + strcpy( output_filename, default_output_filename ); } + if( !open_outstream( force, false ) || !check_tty_out( program_mode ) ) + return 1; /* check tty only once and don't try to delete a tty */ + } + + const struct stat * const in_statsp = + ( input_filename[0] && one_to_one ) ? &in_stats : 0; + int tmp; + if( program_mode == m_compress ) + tmp = compress( infd, &encoder_options, &pp ); + else + tmp = decompress( infd, &pp, ignore_trailing, loose_trailing, + program_mode == m_test ); + if( close( infd ) != 0 ) + { show_file_error( pp.name, "Error closing input file", errno ); + set_retval( &tmp, 1 ); } + set_retval( &retval, tmp ); + if( tmp ) + { if( program_mode != m_test ) cleanup_and_fail( retval ); + else ++failed_tests; } + + if( delete_output_on_interrupt && one_to_one ) + close_and_set_permissions( in_statsp ); + if( input_filename[0] && !keep_input_files && one_to_one ) + remove( input_filename ); + } + if( delete_output_on_interrupt ) /* -o */ + close_and_set_permissions( ( retval == 0 && !stdin_used && + filenames_given && num_filenames == 1 ) ? &in_stats : 0 ); + else if( outfd >= 0 && close( outfd ) != 0 ) /* -c */ + { + show_error( "Error closing stdout", errno, false ); + set_retval( &retval, 1 ); + } + if( failed_tests > 0 && verbosity >= 1 && num_filenames > 1 ) + fprintf( stderr, "%s: warning: %d %s failed the test.\n", + program_name, failed_tests, + ( failed_tests == 1 ) ? "file" : "files" ); + free( output_filename ); + free( filenames ); + ap_free( &parser ); + return retval; + } |