diff options
Diffstat (limited to 'zcmp.cc')
| -rw-r--r-- | zcmp.cc | 148 |
1 files changed, 76 insertions, 72 deletions
@@ -40,7 +40,7 @@ #include "zutils.h" #ifndef LLONG_MAX -#define LLONG_MAX 0x7FFFFFFFFFFFFFFFLL +#define LLONG_MAX 0x7FFFFFFFFFFFFFFFLL #endif @@ -70,27 +70,32 @@ void show_help() " -h, --help display this help and exit\n" " -V, --version output version information and exit\n" " -b, --print-bytes print differing bytes\n" + " -H, --hexadecimal print hexadecimal values instead of octal\n" " -i, --ignore-initial=<n>[:<n2>] ignore differences in the first <n> bytes\n" " -l, --list list position, value of all differing bytes\n" " -M, --format=<list> process only the formats in <list>\n" " -n, --bytes=<n> compare at most <n> bytes\n" " -N, --no-rcfile don't read runtime configuration file\n" - " -O, --force-format=[<f1>][,<f2>] force the formats given (bz2,gz,lz,xz,zst)\n" - " -q, --quiet suppress all messages\n" - " -s, --silent (same as --quiet)\n" - " -v, --verbose verbose mode (same as --list)\n" + " -O, --force-format=[<f1>][,<f2>] force one or both input formats\n" + " -q, --quiet, --silent suppress diagnostics written to stderr\n" + " -s, --script suppress messages about file differences\n" + " -v, --verbose verbose mode (opposite of --quiet)\n" " --bz2=<command> set compressor and options for bzip2 format\n" " --gz=<command> set compressor and options for gzip format\n" " --lz=<command> set compressor and options for lzip format\n" " --xz=<command> set compressor and options for xz format\n" " --zst=<command> set compressor and options for zstd format\n" - "\nNumbers 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" ); + "\nValid formats for options '-M' and '-O' are 'bz2', 'gz', 'lz', 'xz', 'zst',\n" + "and 'un' for uncompressed.\n" + "\nByte counts given as arguments to options may be expressed in decimal,\n" + "hexadecimal, or octal (using the same syntax as integer constants in C++),\n" + "and 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" ); show_help_addr(); } -// separate large numbers >= 100_000 in groups of 3 digits using '_' +// separate numbers of 5 or more digits in groups of 3 digits using '_' const char * format_num3( long long num ) { const char * const si_prefix = "kMGTPEZY"; @@ -103,20 +108,20 @@ const char * format_num3( long long num ) char * p = buf + bufsize - 1; // fill the buffer backwards *p = 0; // terminator const bool negative = num < 0; - if( negative ) num = -num; char prefix = 0; // try binary first, then si - for( int i = 0; i < 8 && num >= 1024 && num % 1024 == 0; ++i ) + for( int i = 0; i < 8 && num != 0 && ( num / 1024 ) * 1024 == num; ++i ) { num /= 1024; prefix = binary_prefix[i]; } if( prefix ) *(--p) = 'i'; else - for( int i = 0; i < 8 && num >= 1000 && num % 1000 == 0; ++i ) + for( int i = 0; i < 8 && num != 0 && ( num / 1000 ) * 1000 == num; ++i ) { num /= 1000; prefix = si_prefix[i]; } if( prefix ) *(--p) = prefix; - const bool split = num >= 100000; + const bool split = num >= 10000 || num <= -10000; for( int i = 0; ; ) { - *(--p) = num % 10 + '0'; num /= 10; if( num == 0 ) break; + long long onum = num; num /= 10; + *(--p) = llabs( onum - ( 10 * num ) ) + '0'; if( num == 0 ) break; if( split && ++i >= 3 ) { i = 0; *(--p) = '_'; } } if( negative ) *(--p) = '-'; @@ -133,12 +138,8 @@ long long getnum( const char * const arg, const char * const option_name, errno = 0; long long result = strtoll( arg, &tail, 0 ); if( tail == arg ) - { - if( verbosity >= 0 ) - std::fprintf( stderr, "%s: Bad or missing numerical argument in " - "option '%s'.\n", program_name, option_name ); - std::exit( 2 ); - } + { show_option_error( arg, "Bad or missing numerical argument in", + option_name ); std::exit( 2 ); } if( result < 0 ) errno = ERANGE; if( !errno && tail[0] && std::isalpha( tail[0] ) ) @@ -163,12 +164,8 @@ long long getnum( const char * const arg, const char * const option_name, case 'B': if( factor == 1000 && !bsuf ) exponent = 0; break; } if( exponent < 0 ) - { - if( verbosity >= 0 ) - std::fprintf( stderr, "%s: Bad multiplier in numerical argument of " - "option '%s'.\n", program_name, option_name ); - std::exit( 2 ); - } + { show_option_error( arg, "Bad multiplier in numerical argument of", + option_name ); std::exit( 2 ); } for( int i = 0; i < exponent; ++i ) { if( ulimit / factor >= result ) result *= factor; @@ -179,8 +176,8 @@ long long getnum( const char * const arg, const char * const option_name, if( errno ) { if( verbosity >= 0 ) - std::fprintf( stderr, "%s: Numerical argument out of limits [%s,%s] " - "in option '%s'.\n", program_name, format_num3( llimit ), + std::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 ); std::exit( 2 ); } @@ -197,13 +194,7 @@ void parse_ignore_initial( const char * const arg, const char * const pn, if( *tail == ':' || *tail == ',' ) ignore_initial[1] = getnum( ++tail, pn ); else if( *tail == 0 ) ignore_initial[1] = ignore_initial[0]; - else - { - if( verbosity >= 0 ) - std::fprintf( stderr, "%s: Bad separator in argument of option '%s'.\n", - program_name, pn ); - std::exit( 2 ); - } + else { show_option_error( arg, "Missing colon in", pn ); std::exit( 2 ); } } @@ -227,8 +218,8 @@ bool skip_ignore_initial( const long long ignore_initial, const int infd ) } -// Put into buf the unsigned char c, making unprintable bytes -// visible by quoting like cat -t does. +/* Put into buf the unsigned char c, making unprintable bytes visible by + quoting like cat -t does. */ void sprintc( char * const buf, unsigned char c ) { int i = 0; @@ -251,7 +242,7 @@ int block_compare( const uint8_t * const buffer0, const uint8_t * p0 = buffer0; const uint8_t * p1 = buffer1; - if( verbosity == 0 ) + if( line_numberp ) { int nl_count = 0; while( *p0 == *p1 ) @@ -265,7 +256,8 @@ int block_compare( const uint8_t * const buffer0, int cmp( const long long max_size, const int infd[2], const std::string filenames[2], bool finished[2], - const bool print_bytes ) + const bool hexadecimal, const bool list, const bool print_bytes, + const bool scripted ) { const int buffer_size = 4096; unsigned long long byte_number = 1; @@ -277,7 +269,7 @@ int cmp( const long long max_size, const int infd[2], uint8_t * const buffer1 = buffer0 + buffer_size + 1; uint8_t * buffer[2]; buffer[0] = buffer0; buffer[1] = buffer1; - int different = 0; + int retval = 0; while( rest > 0 ) { @@ -289,7 +281,7 @@ int cmp( const long long max_size, const int infd[2], rd[i] = readblock( infd[i], buffer[i], size ); if( rd[i] != size && errno ) { show_file_error( filenames[i].c_str(), "Read error", errno ); - return 2; } + retval = 2; goto done; } } for( int i = 0; i < 2; ++i ) if( rd[i] < size ) finished[i] = true; @@ -298,13 +290,14 @@ int cmp( const long long max_size, const int infd[2], buffer0[min_rd] = 0; // sentinels for the block compare buffer1[min_rd] = 1; - int first_diff = block_compare( buffer0, buffer1, &line_number ); + int first_diff = block_compare( buffer0, buffer1, list ? 0 : &line_number ); byte_number += first_diff; if( first_diff < min_rd ) { - if( verbosity < 0 ) return 1; // return status only - if( verbosity == 0 ) // show first difference + retval = 1; // difference found + if( scripted ) break; // status only + if( !list ) // show first difference { if( !print_bytes ) std::printf( "%s %s differ: byte %llu, line %llu\n", @@ -316,16 +309,17 @@ int cmp( const long long max_size, const int infd[2], const unsigned char c1 = buffer1[first_diff]; char buf0[5], buf1[5]; sprintc( buf0, c0 ); sprintc( buf1, c1 ); - std::printf( "%s %s differ: byte %llu, line %llu is %3o %s %3o %s\n", + std::printf( hexadecimal ? + "%s %s differ: byte %llu, line %llu is %02X %s %02X %s\n" : + "%s %s differ: byte %llu, line %llu is %3o %s %3o %s\n", filenames[0].c_str(), filenames[1].c_str(), byte_number, line_number, c0, buf0, c1, buf1 ); } std::fflush( stdout ); - return 1; + break; } - else // verbosity > 0 ; show all differences + else // list ; show all differences { - different = 1; for( ; first_diff < min_rd; ++byte_number, ++first_diff ) { const unsigned char c0 = buffer0[first_diff]; @@ -333,12 +327,14 @@ int cmp( const long long max_size, const int infd[2], if( c0 != c1 ) { if( !print_bytes ) - std::printf( "%llu %3o %3o\n", byte_number, c0, c1 ); + std::printf( hexadecimal ? "%llu %02X %02X\n" : "%llu %3o %3o\n", + byte_number, c0, c1 ); else { char buf0[5], buf1[5]; sprintc( buf0, c0 ); sprintc( buf1, c1 ); - std::printf( "%llu %3o %-4s %3o %s\n", + std::printf( hexadecimal ? "%llu %02X %-4s %02X %s\n" : + "%llu %3o %-4s %3o %s\n", byte_number, c0, buf0, c1, buf1 ); } } @@ -350,15 +346,18 @@ int cmp( const long long max_size, const int infd[2], if( rd[0] != rd[1] ) { if( verbosity >= 0 ) - std::fprintf( stderr, "%s: EOF on %s\n", - program_name, filenames[rd[1]<rd[0]].c_str() ); - return 1; + std::fprintf( stderr, list ? + "%s: EOF on %s after byte %llu\n" : + "%s: EOF on %s after byte %llu, in line %llu\n", + program_name, filenames[rd[1]<rd[0]].c_str(), + byte_number - 1, line_number ); + retval = 1; break; } if( min_rd != buffer_size ) break; } - +done: delete[] buffer0; - return different; + return retval; } } // end namespace @@ -370,8 +369,11 @@ int main( const int argc, const char * const argv[] ) // number of initial bytes ignored for each file long long ignore_initial[2] = { 0, 0 }; long long max_size = -1; // < 0 means unlimited size - int format_types[2] = { -1, -1 }; - bool print_bytes = false; + int format_types[2] = { -1, -1 }; // < 0 means undefined + bool hexadecimal = false; + bool list = false; // list position, value of all differing bytes + bool print_bytes = false; // print differing bytes + bool scripted = false; // suppress messages about file differences program_name = "zcmp"; invocation_name = ( argc > 0 ) ? argv[0] : program_name; @@ -379,6 +381,7 @@ int main( const int argc, const char * const argv[] ) { { 'b', "print-bytes", Arg_parser::no }, { 'h', "help", Arg_parser::no }, + { 'H', "hexadecimal", Arg_parser::no }, { 'i', "ignore-initial", Arg_parser::yes }, { 'l', "list", Arg_parser::no }, { 'M', "format", Arg_parser::yes }, @@ -386,7 +389,8 @@ int main( const int argc, const char * const argv[] ) { 'N', "no-rcfile", Arg_parser::no }, { 'O', "force-format", Arg_parser::yes }, { 'q', "quiet", Arg_parser::no }, - { 's', "silent", Arg_parser::no }, + { 'q', "silent", Arg_parser::no }, + { 's', "script", Arg_parser::no }, { 'v', "verbose", Arg_parser::no }, { 'V', "version", Arg_parser::no }, { bz2_opt, "bz2", Arg_parser::yes }, @@ -414,21 +418,22 @@ int main( const int argc, const char * const argv[] ) { case 'b': print_bytes = true; break; case 'h': show_help(); return 0; + case 'H': hexadecimal = true; break; case 'i': parse_ignore_initial( arg, pn, ignore_initial ); break; - case 'l': verbosity = 1; break; + case 'l': list = true; break; case 'M': parse_format_list( sarg, pn ); break; case 'n': max_size = getnum( arg, pn ); break; case 'N': break; case 'O': parse_format_types2( sarg, pn, format_types ); break; - case 'q': - case 's': verbosity = -1; break; - case 'v': verbosity = 1; break; + case 'q': verbosity = -1; break; + case 's': scripted = true; break; + case 'v': if( verbosity < 4 ) ++verbosity; break; case 'V': show_version(); return 0; - case bz2_opt: parse_compressor( sarg, fmt_bz2 ); break; - case gz_opt: parse_compressor( sarg, fmt_gz ); break; - case lz_opt: parse_compressor( sarg, fmt_lz ); break; - case xz_opt: parse_compressor( sarg, fmt_xz ); break; - case zst_opt: parse_compressor( sarg, fmt_zst ); break; + case bz2_opt: parse_compressor( sarg, pn, fmt_bz2 ); break; + case gz_opt: parse_compressor( sarg, pn, fmt_gz ); break; + case lz_opt: parse_compressor( sarg, pn, fmt_lz ); break; + case xz_opt: parse_compressor( sarg, pn, fmt_xz ); break; + case zst_opt: parse_compressor( sarg, pn, fmt_zst ); break; default : internal_error( "uncaught option." ); } } // end process options @@ -438,12 +443,10 @@ int main( const int argc, const char * const argv[] ) setmode( STDOUT_FILENO, O_BINARY ); #endif - if( argind >= parser.arguments() ) - { show_error( "No files given.", 0, true ); return 2; } - if( parser.arguments() - argind > 2 ) - { show_error( "Too many files.", 0, true ); return 2; } - const int files = parser.arguments() - argind; + if( files < 1 ) { show_error( "No files given.", 0, true ); return 2; } + if( files > 2 ) { show_error( "Too many files.", 0, true ); return 2; } + std::string filenames[2]; // file names of the two input files filenames[0] = parser.argument( argind ); if( files == 2 ) filenames[1] = parser.argument( argind + 1 ); @@ -498,7 +501,8 @@ int main( const int argc, const char * const argv[] ) } bool finished[2] = { false, false }; - int retval = cmp( max_size, infd, filenames, finished, print_bytes ); + int retval = cmp( max_size, infd, filenames, finished, hexadecimal, list, + print_bytes, scripted ); for( int i = 0; i < 2; ++i ) if( !good_status( children[i], finished[i] ) ) retval = 2; |