summaryrefslogtreecommitdiffstats
path: root/main.cc
diff options
context:
space:
mode:
Diffstat (limited to '')
-rw-r--r--main.cc275
1 files changed, 150 insertions, 125 deletions
diff --git a/main.cc b/main.cc
index d3ef70c..95ee71b 100644
--- a/main.cc
+++ b/main.cc
@@ -1,5 +1,5 @@
/* Lziprecover - Data recovery tool for the lzip format
- Copyright (C) 2009-2016 Antonio Diaz Diaz.
+ Copyright (C) 2009-2017 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
@@ -73,10 +73,10 @@ namespace {
const char * const Program_name = "Lziprecover";
const char * const program_name = "lziprecover";
-const char * const program_year = "2016";
+const char * const program_year = "2017";
const char * invocation_name = 0;
-struct { const char * from; const char * to; } const known_extensions[] = {
+const struct { const char * from; const char * to; } known_extensions[] = {
{ ".lz", "" },
{ ".tlz", ".tar" },
{ 0, 0 } };
@@ -99,6 +99,8 @@ void show_help()
"\nLziprecover can also produce a correct file by merging the good parts of\n"
"two or more damaged copies, extract data from damaged files, decompress\n"
"files and test integrity of files.\n"
+ "\nLziprecover provides random access to the data in multimember files; it\n"
+ "only decompresses the members containing the desired data.\n"
"\nLziprecover is not a replacement for regular backups, but a last line of\n"
"defense for the case where the backups are also damaged.\n"
"\nUsage: %s [options] [files]\n", invocation_name );
@@ -113,7 +115,7 @@ void show_help()
" -f, --force overwrite existing output files\n"
" -i, --ignore-errors make '--range-decompress' ignore data errors\n"
" -k, --keep keep (don't delete) input files\n"
- " -l, --list print total file sizes and ratios\n"
+ " -l, --list print (un)compressed file sizes\n"
" -m, --merge correct errors in file using several copies\n"
" -o, --output=<file> place the output into <file>\n"
" -q, --quiet suppress all messages\n"
@@ -152,42 +154,64 @@ void show_version()
} // end namespace
+const char * bad_version( const unsigned version )
+ {
+ static char buf[80];
+ snprintf( buf, sizeof buf, "Version %u member format not supported.",
+ version );
+ return buf;
+ }
+
+
+const char * format_ds( const unsigned dictionary_size )
+ {
+ enum { bufsize = 16, factor = 1024 };
+ static char buf[bufsize];
+ const char * const prefix[8] =
+ { "Ki", "Mi", "Gi", "Ti", "Pi", "Ei", "Zi", "Yi" };
+ const char * p = "";
+ const char * np = " ";
+ unsigned num = dictionary_size;
+ bool exact = ( num % factor == 0 );
+
+ for( int i = 0; i < 8 && ( num > 9999 || ( exact && num >= factor ) ); ++i )
+ { num /= factor; if( num % factor != 0 ) exact = false;
+ p = prefix[i]; np = ""; }
+ snprintf( buf, bufsize, "%s%4u %sB", np, num, p );
+ return buf;
+ }
+
+
void show_header( const unsigned dictionary_size, const int vlevel )
{
if( verbosity >= vlevel )
- {
- const char * const prefix[8] =
- { "Ki", "Mi", "Gi", "Ti", "Pi", "Ei", "Zi", "Yi" };
- enum { factor = 1024 };
- const char * p = "";
- const char * np = " ";
- unsigned num = dictionary_size;
- bool exact = ( num % factor == 0 );
-
- for( int i = 0; i < 8 && ( num > 9999 || ( exact && num >= factor ) ); ++i )
- { num /= factor; if( num % factor != 0 ) exact = false;
- p = prefix[i]; np = ""; }
- std::fprintf( stderr, "dictionary size %s%4u %sB. ", np, num, p );
- }
+ std::fprintf( stderr, "dictionary %s. ", format_ds( dictionary_size ) );
}
namespace {
-// Returns the number of chars read, or 0 if error.
-//
-int parse_long_long( const char * const ptr, long long & value )
+long long getnum( const char * const ptr, const long long llimit = -LLONG_MAX,
+ const long long ulimit = LLONG_MAX,
+ const char ** const tailp = 0 )
{
char * tail;
errno = 0;
- value = strtoll( ptr, &tail, 0 );
- if( tail == ptr || errno || value < 0 ) return 0;
- int c = tail - ptr;
+ long long result = strtoll( ptr, &tail, 0 );
+ if( tail == ptr )
+ {
+ show_error( "Bad or missing numerical argument.", 0, true );
+ std::exit( 1 );
+ }
- if( ptr[c] )
+ if( !errno && tail[0] )
{
- const int factor = ( ptr[c+1] == 'i' ) ? 1024 : 1000;
- int exponent = 0;
- switch( ptr[c] )
+ char * const p = tail++;
+ int factor;
+ bool bsuf; // 'B' suffix is present
+ if( tail[0] == 'i' ) { ++tail; factor = 1024; } else factor = 1000;
+ if( tail[0] == 'B' ) { ++tail; bsuf = true; } else bsuf = false;
+ int exponent = -1; // -1 = bad multiplier
+ switch( *p )
{
case 'Y': exponent = 8; break;
case 'Z': exponent = 7; break;
@@ -196,22 +220,30 @@ int parse_long_long( const char * const ptr, long long & value )
case 'T': exponent = 4; break;
case 'G': exponent = 3; break;
case 'M': exponent = 2; break;
- case 'K': if( factor == 1024 ) exponent = 1; else return 0; break;
- case 'k': if( factor == 1000 ) exponent = 1; else return 0; break;
+ case 'K': if( factor == 1024 ) exponent = 1; break;
+ case 'k': if( factor == 1000 ) exponent = 1; break;
+ case 'B': if( factor == 1000 && !bsuf ) exponent = 0; break;
+ default : if( tailp ) { tail = p; exponent = 0; } break;
}
- if( exponent > 0 )
+ if( exponent < 0 )
{
- ++c;
- if( ptr[c] == 'i' ) { ++c; if( value ) format_num( 0, 0, -1 ); }
- if( ptr[c] == 'B' ) ++c;
- for( int i = 0; i < exponent; ++i )
- {
- if( INT64_MAX / factor >= value ) value *= factor;
- else return 0;
- }
+ show_error( "Bad multiplier in numerical argument.", 0, true );
+ std::exit( 1 );
+ }
+ for( int i = 0; i < exponent; ++i )
+ {
+ if( LLONG_MAX / factor >= std::labs( result ) ) result *= factor;
+ else { errno = ERANGE; break; }
}
}
- return c;
+ if( !errno && ( result < llimit || result > ulimit ) ) errno = ERANGE;
+ if( errno )
+ {
+ show_error( "Numerical argument out of limits." );
+ std::exit( 1 );
+ }
+ if( tailp ) *tailp = tail;
+ return result;
}
@@ -219,17 +251,16 @@ int parse_long_long( const char * const ptr, long long & value )
//
void parse_range( const char * const ptr, Block & range )
{
- long long value = 0;
- const bool size_only = ( ptr[0] == ',' );
- int c = size_only ? 0 : parse_long_long( ptr, value ); // pos
- if( size_only || ( c && value >= 0 && value < INT64_MAX &&
- ( ptr[c] == 0 || ptr[c] == ',' || ptr[c] == '-' ) ) )
+ const char * tail = ptr;
+ long long value =
+ ( ptr[0] == ',' ) ? 0 : getnum( ptr, 0, INT64_MAX - 1, &tail );
+ if( tail[0] == 0 || tail[0] == ',' || tail[0] == '-' )
{
range.pos( value );
- 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( tail[0] == 0 ) { range.size( INT64_MAX - value ); return; }
+ const bool issize = ( tail[0] == ',' );
+ value = getnum( tail + 1, 1, INT64_MAX ); // size
+ if( issize || value > range.pos() )
{
if( !issize ) value -= range.pos();
if( INT64_MAX - range.pos() >= value ) { range.size( value ); return; }
@@ -240,21 +271,21 @@ void parse_range( const char * const ptr, Block & range )
}
-// Recognized format: <pos>,<value>
+// Recognized formats: <pos>,<value> <pos>,+<value> <pos>,f<value>
//
-void parse_pos_value( const char * const ptr, long long & pos, uint8_t & value )
+void parse_pos_value( const char * const ptr, Bad_byte & bad_byte )
{
- long long val = 0;
- int c = parse_long_long( ptr, val ); // pos
- if( c && val >= 0 && val < INT64_MAX && ptr[c] == ',' )
+ const char * tail;
+ bad_byte.pos = getnum( ptr, 0, INT64_MAX, &tail );
+ if( tail[0] != ',' )
{
- pos = val;
- c = parse_long_long( ptr + c + 1, val ); // value
- if( c && val >= 0 && val < 256 )
- { value = val; return; }
+ show_error( "Bad separator between <pos> and <val>.", 0, true );
+ std::exit( 1 );
}
- show_error( "Bad file position or byte value.", 0, true );
- std::exit( 1 );
+ if( tail[1] == '+' ) { ++tail; bad_byte.mode = Bad_byte::delta; }
+ else if( tail[1] == 'f' ) { ++tail; bad_byte.mode = Bad_byte::flip; }
+ else bad_byte.mode = Bad_byte::literal;
+ bad_byte.value = getnum( tail + 1, 0, 255 );
}
@@ -281,12 +312,12 @@ void set_mode( Mode & program_mode, const Mode new_mode )
int extension_index( const std::string & name )
{
- for( int i = 0; known_extensions[i].from; ++i )
+ for( int eindex = 0; known_extensions[eindex].from; ++eindex )
{
- const std::string ext( known_extensions[i].from );
+ const std::string ext( known_extensions[eindex].from );
if( name.size() > ext.size() &&
name.compare( name.size() - ext.size(), ext.size(), ext ) == 0 )
- return i;
+ return eindex;
}
return -1;
}
@@ -298,11 +329,7 @@ int open_instream( const char * const name, struct stat * const in_statsp,
{
int 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 ) );
- }
+ show_file_error( name, "Can't open input file", errno );
else
{
const int i = fstat( infd, in_statsp );
@@ -338,15 +365,15 @@ void set_a_outname( const std::string & name )
}
-void set_d_outname( const std::string & name, const int i )
+void set_d_outname( const std::string & name, const int eindex )
{
- if( i >= 0 )
+ if( eindex >= 0 )
{
- const std::string from( known_extensions[i].from );
+ const std::string from( known_extensions[eindex].from );
if( name.size() > from.size() )
{
output_filename.assign( name, 0, name.size() - from.size() );
- output_filename += known_extensions[i].to;
+ output_filename += known_extensions[eindex].to;
return;
}
}
@@ -398,7 +425,8 @@ bool file_exists( const std::string & filename )
}
-bool check_tty( const int infd, const Mode program_mode )
+bool check_tty( const char * const input_filename, const int infd,
+ const Mode program_mode )
{
if( program_mode == m_alone_to_lz && isatty( outfd ) )
{
@@ -407,7 +435,8 @@ bool check_tty( const int infd, const Mode program_mode )
}
if( isatty( infd ) ) // all modes read compressed data
{
- show_error( "I won't read compressed data from a terminal.", 0, true );
+ show_file_error( input_filename,
+ "I won't read compressed data from a terminal." );
return false;
}
return true;
@@ -465,10 +494,10 @@ void close_and_set_permissions( const struct stat * const in_statsp )
}
-unsigned char xdigit( const int value )
+unsigned char xdigit( const unsigned value )
{
- if( value >= 0 && value <= 9 ) return '0' + value;
- if( value >= 10 && value <= 15 ) return 'A' + value - 10;
+ if( value <= 9 ) return '0' + value;
+ if( value <= 15 ) return 'A' + value - 10;
return 0;
}
@@ -482,26 +511,18 @@ bool show_trailing_data( const uint8_t * const data, const int size,
std::string msg;
if( !all ) msg = "first bytes of ";
msg += "trailing data = ";
- bool text = true;
for( int i = 0; i < size; ++i )
- if( !std::isprint( data[i] ) ) { text = false; break; }
- if( text )
- {
- msg += '\'';
- msg.append( (const char *)data, size );
- msg += '\'';
- }
- else
{
- for( int i = 0; i < size; ++i )
- {
- if( i > 0 ) msg += ' ';
- msg += xdigit( data[i] >> 4 );
- msg += xdigit( data[i] & 0x0F );
- }
+ msg += xdigit( data[i] >> 4 );
+ msg += xdigit( data[i] & 0x0F );
+ msg += ' ';
}
+ msg += '\'';
+ for( int i = 0; i < size; ++i )
+ { if( std::isprint( data[i] ) ) msg += data[i]; else msg += '.'; }
+ msg += '\'';
pp( msg.c_str() );
- if( !ignore_trailing ) show_error( "Trailing data not allowed." );
+ if( !ignore_trailing ) show_file_error( pp.name(), trailing_msg );
}
return ignore_trailing;
}
@@ -532,22 +553,16 @@ int decompress( const int infd, const Pretty_print & pp,
if( !header.verify_magic() )
{
if( first_member )
- { pp( "Bad magic number (file not in lzip format)." ); retval = 2; }
+ { show_file_error( pp.name(), bad_magic_msg ); retval = 2; }
else if( !show_trailing_data( header.data, size, pp, false, ignore_trailing ) )
retval = 2;
break;
}
if( !header.verify_version() )
- {
- if( verbosity >= 0 )
- { pp();
- std::fprintf( stderr, "Version %d member format not supported.\n",
- header.version() ); }
- retval = 2; break;
- }
+ { pp( bad_version( header.version() ) ); retval = 2; break; }
const unsigned dictionary_size = header.dictionary_size();
if( !isvalid_ds( dictionary_size ) )
- { pp( "Invalid dictionary size in member header." ); retval = 2; break; }
+ { pp( bad_dict_msg ); retval = 2; break; }
if( verbosity >= 2 || ( verbosity == 1 && first_member ) )
{ pp(); show_header( dictionary_size ); }
@@ -634,6 +649,16 @@ void show_error( const char * const msg, const int errcode, const bool help )
}
+void show_file_error( const char * const filename, const char * const msg,
+ const int errcode )
+ {
+ if( verbosity < 0 ) return;
+ std::fprintf( stderr, "%s: %s: %s", program_name, filename, msg );
+ if( errcode > 0 ) std::fprintf( stderr, ": %s", std::strerror( errcode ) );
+ std::fputc( '\n', stderr );
+ }
+
+
void internal_error( const char * const msg )
{
if( verbosity >= 0 )
@@ -662,13 +687,11 @@ void show_error4( const char * const msg1, const char * const name1,
int main( const int argc, const char * const argv[] )
{
Block range( 0, 0 );
- long long bad_pos = -1;
- std::string input_filename;
+ Bad_byte bad_byte;
std::string default_output_filename;
std::vector< std::string > filenames;
int infd = -1;
Mode program_mode = m_none;
- uint8_t bad_value = 0;
bool force = false;
bool ignore_errors = false;
bool ignore_trailing = true;
@@ -712,8 +735,8 @@ int main( const int argc, const char * const argv[] )
{
const int code = parser.code( argind );
if( !code ) break; // no more options
- const std::string & arg = parser.argument( argind );
- const char * const ptr = arg.c_str();
+ const std::string & sarg = parser.argument( argind );
+ const char * const arg = sarg.c_str();
switch( code )
{
case 'a': ignore_trailing = false; break;
@@ -721,7 +744,7 @@ int main( const int argc, const char * const argv[] )
case 'c': to_stdout = true; break;
case 'd': set_mode( program_mode, m_decompress ); break;
case 'D': set_mode( program_mode, m_range_dec );
- parse_range( ptr, range ); break;
+ parse_range( arg, range ); break;
case 'f': force = true; break;
case 'h': show_help(); return 0;
case 'i': ignore_errors = true; break;
@@ -729,7 +752,7 @@ int main( const int argc, const char * const argv[] )
case 'l': set_mode( program_mode, m_list ); break;
case 'm': set_mode( program_mode, m_merge ); break;
case 'n': break;
- case 'o': default_output_filename = arg; break;
+ case 'o': default_output_filename = sarg; break;
case 'q': verbosity = -1; break;
case 'R': set_mode( program_mode, m_repair ); break;
case 's': set_mode( program_mode, m_split ); break;
@@ -737,13 +760,13 @@ int main( const int argc, const char * const argv[] )
case 'v': if( verbosity < 4 ) ++verbosity; break;
case 'V': show_version(); return 0;
case 'W': set_mode( program_mode, m_debug_decompress );
- parse_pos_value( ptr, bad_pos, bad_value ); break;
+ parse_pos_value( arg, bad_byte ); break;
case 'X': set_mode( program_mode, m_show_packets );
- if( ptr[0] ) parse_pos_value( ptr, bad_pos, bad_value ); break;
+ if( arg[0] ) { parse_pos_value( arg, bad_byte ); } break;
case 'Y': set_mode( program_mode, m_debug_delay );
- parse_range( ptr, range ); break;
+ parse_range( arg, range ); break;
case 'Z': set_mode( program_mode, m_debug_repair );
- parse_pos_value( ptr, bad_pos, bad_value ); break;
+ parse_pos_value( arg, bad_byte ); break;
default : internal_error( "uncaught option." );
}
} // end process options
@@ -773,18 +796,15 @@ int main( const int argc, const char * const argv[] )
case m_alone_to_lz: break;
case m_debug_decompress:
one_file( filenames.size() );
- return debug_decompress( filenames[0], bad_pos, verbosity, bad_value, false );
+ return debug_decompress( filenames[0], bad_byte, verbosity, false );
case m_debug_delay:
one_file( filenames.size() );
return debug_delay( filenames[0], range, verbosity );
case m_debug_repair:
one_file( filenames.size() );
- return debug_repair( filenames[0], bad_pos, verbosity, bad_value );
+ return debug_repair( filenames[0], bad_byte, verbosity );
case m_decompress: break;
- case m_list:
- if( filenames.size() < 1 )
- { show_error( "You must specify at least 1 file.", 0, true ); return 1; }
- return list_files( filenames, verbosity );
+ case m_list: break;
case m_merge:
if( filenames.size() < 2 )
{ show_error( "You must specify at least 2 files.", 0, true ); return 1; }
@@ -794,7 +814,7 @@ int main( const int argc, const char * const argv[] )
one_file( filenames.size() );
set_signals();
return range_decompress( filenames[0], default_output_filename, range,
- verbosity, force, ignore_errors, to_stdout );
+ verbosity, force, ignore_errors, ignore_trailing, to_stdout );
case m_repair:
one_file( filenames.size() );
set_signals();
@@ -802,7 +822,7 @@ int main( const int argc, const char * const argv[] )
force );
case m_show_packets:
one_file( filenames.size() );
- return debug_decompress( filenames[0], bad_pos, verbosity, bad_value, true );
+ return debug_decompress( filenames[0], bad_byte, verbosity, true );
case m_split:
one_file( filenames.size() );
set_signals();
@@ -814,12 +834,16 @@ int main( const int argc, const char * const argv[] )
{ show_error( "Not enough memory." ); cleanup_and_fail( 1 ); }
catch( Error e ) { show_error( e.msg, errno ); cleanup_and_fail( 1 ); }
+ if( filenames.empty() ) filenames.push_back("-");
+
+ if( program_mode == m_list )
+ return list_files( filenames, verbosity, ignore_trailing );
+
if( program_mode == m_test )
outfd = -1;
else if( program_mode != m_alone_to_lz && program_mode != m_decompress )
internal_error( "invalid decompressor operation." );
- if( filenames.empty() ) filenames.push_back("-");
if( !to_stdout && program_mode != m_test &&
( filenames_given || default_output_filename.size() ) )
set_signals();
@@ -830,13 +854,13 @@ int main( const int argc, const char * const argv[] )
bool stdin_used = false;
for( unsigned i = 0; i < filenames.size(); ++i )
{
+ std::string input_filename;
struct stat in_stats;
output_filename.clear();
if( filenames[i].empty() || filenames[i] == "-" )
{
if( stdin_used ) continue; else stdin_used = true;
- input_filename.clear();
infd = STDIN_FILENO;
if( program_mode != m_test )
{
@@ -881,14 +905,15 @@ int main( const int argc, const char * const argv[] )
}
}
- if( !check_tty( infd, program_mode ) )
+ pp.set_name( input_filename );
+ if( !check_tty( pp.name(), infd, program_mode ) )
{
if( retval < 1 ) retval = 1;
+ if( program_mode == m_test ) { close( infd ); infd = -1; continue; }
cleanup_and_fail( retval );
}
const struct stat * const in_statsp = input_filename.size() ? &in_stats : 0;
- pp.set_name( input_filename );
int tmp;
if( program_mode == m_alone_to_lz )
tmp = alone_to_lz( infd, pp );