lziprecover/testsuite/unzcrash.cc

/*  Unzcrash - Tests robustness of decompressors to corrupted data.
    Inspired by unzcrash.c from Julian Seward's bzip2.
    Copyright (C) 2008-2014 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 2 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 <http://www.gnu.org/licenses/>.
*/
/*
    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 unzcrash to panic.
*/

#include <cerrno>
#include <climits>
#include <csignal>
#include <cstdio>
#include <cstdlib>
#include <cstring>
#include <string>
#include <vector>
#include <stdint.h>
#include <unistd.h>

#include "../arg_parser.h"

#if CHAR_BIT != 8
#error "Environments where CHAR_BIT != 8 are not supported."
#endif


namespace {

const char * const Program_name = "Unzcrash";
const char * const program_name = "unzcrash";
const char * const program_year = "2014";
const char * invocation_name = 0;

int verbosity = 0;


void show_help()
  {
  std::printf( "%s - Tests robustness of decompressors to corrupted data.\n", Program_name );
  std::printf( "\nUsage: %s [options] \"lzip -tv\" filename.lz\n", invocation_name );
  std::printf( "\nThis program reads the specified file and then repeatedly decompresses\n"
               "it, increasing 256 times each byte of the compressed data, so as to test\n"
               "all possible one-byte errors. This should not cause any invalid memory\n"
               "accesses. If it does, please, report it as a bug.\n"
               "\nOptions:\n"
               "  -h, --help                 display this help and exit\n"
               "  -V, --version              output version information and exit\n"
               "  -b, --bits=<range>         test N-bit errors instead of full byte\n"
               "  -p, --position=<bytes>     first byte position to test [default 0]\n"
               "  -q, --quiet                suppress all messages\n"
               "  -s, --size=<bytes>         number of byte positions to test [all]\n"
               "  -v, --verbose              be verbose (a 2nd -v gives more)\n"
               "Examples of <range>:  1  1,2,3  1-4  1,3-5,8  1-3,5-8\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 unzcrash to panic.\n"
               "\nReport bugs to lzip-bug@nongnu.org\n"
               "Lziprecover home page: http://www.nongnu.org/lzip/lziprecover.html\n" );
  }


void show_version()
  {
  std::printf( "%s %s\n", program_name, PROGVERSION );
  std::printf( "Copyright (C) %s Antonio Diaz Diaz.\n", program_year );
  std::printf( "License GPLv2+: GNU GPL version 2 or later <http://gnu.org/licenses/gpl.html>\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" );
  }


void show_error( const char * const msg, const int errcode = 0,
                 const bool help = false )
  {
  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 );
  }


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;
  }


class Bitset8			// 8 value bitset (1 to 8)
  {
  bool data[8];
  static bool valid_digit( const unsigned char ch )
    { return ( ch >= '1' && ch <= '8' ); }

public:
  Bitset8() { for( int i = 0; i < 8; ++i ) data[i] = true; }

  bool includes( const int i ) const
    { return ( i >= 1 && i <= 8 && data[i-1] ); }

  // Recognized formats: 1 1,2,3 1-4 1,3-5,8 1-3,5-8
  bool parse( const char * p )
    {
    for( int i = 0; i < 8; ++i ) data[i] = false;
    while( true )
      {
      const unsigned char ch1 = *p++;
      if( !valid_digit( ch1 ) ) break;
      if( *p != '-' ) data[ch1-'1'] = true;
      else
        {
        ++p;
        if( !valid_digit( *p ) || ch1 > *p ) break;
        for( int c = ch1; c <= *p; ++c ) data[c-'1'] = true;
        ++p;
        }
      if( *p == 0 ) return true;
      if( *p == ',' ) ++p; else break;
      }
    show_error( "Invalid value or range." );
    return false;
    }

  // number of N-bit errors per byte (N=0 to 8): 1 8 28 56 70 56 28 8 1
  void print() const
    {
    std::fflush( stderr );
    int c = 0;
    for( int i = 0; i < 8; ++i ) if( data[i] ) ++c;
    if( c == 8 ) std::printf( "Testing full byte.\n" );
    else if( c == 0 ) std::printf( "Nothing to test.\n" );
    else
      {
      std::printf( "Testing " );
      for( int i = 0; i < 8; ++i )
        if( data[i] )
          {
          std::printf( "%d", i + 1 );
          if( --c ) std::printf( "," );
          }
      std::printf( " bit errors.\n" );
      }
    std::fflush( stdout );
    }
  };


int differing_bits( const uint8_t byte1, const uint8_t byte2 )
  {
  int count = 0;
  uint8_t dif = byte1 ^ byte2;
  while( dif )
    { count += ( dif & 1 ); dif >>= 1; }
  return count;
  }

} // end namespace


int main( const int argc, const char * const argv[] )
  {
  enum { buffer_size = 3 << 20 };
  Bitset8 bits;			// if Bitset8::parse not called test full byte
  int pos = 0;
  int max_size = buffer_size;
  invocation_name = argv[0];

  const Arg_parser::Option options[] =
    {
    { 'h', "help",     Arg_parser::no  },
    { 'b', "bits",     Arg_parser::yes },
    { 'p', "position", Arg_parser::yes },
    { 'q', "quiet",    Arg_parser::no  },
    { 's', "size",     Arg_parser::yes },
    { '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 'h': show_help(); return 0;
      case 'b': if( !bits.parse( arg ) ) return 1; break;
      case 'p': pos = getnum( arg, 0, buffer_size - 1 ); break;
      case 'q': verbosity = -1; break;
      case 's': max_size = getnum( arg, 1, buffer_size ); break;
      case 'v': if( verbosity < 4 ) ++verbosity; break;
      case 'V': show_version(); return 0;
      default : internal_error( "uncaught option." );
      }
    } // end process options

  if( argind + 2 != parser.arguments() )
    {
    if( verbosity >= 0 )
      std::fprintf( stderr, "Usage: %s \"lzip -tv\" filename.lz\n",
                    invocation_name );
    return 1;
    }

  FILE *f = std::fopen( parser.argument( argind + 1 ).c_str(), "rb" );
  if( !f )
    {
    if( verbosity >= 0 )
      std::fprintf( stderr, "Can't open file '%s' for reading\n",
                    parser.argument( argind + 1 ).c_str() );
    return 1;
    }

  uint8_t * const buffer = new uint8_t[buffer_size];
  const int size = std::fread( buffer, 1, buffer_size, f );
  if( size >= buffer_size )
    {
    if( verbosity >= 0 )
      std::fprintf( stderr, "input file '%s' is too large.\n",
                    parser.argument( argind + 1 ).c_str() );
    return 2;
    }
  std::fclose( f );

  f = popen( parser.argument( argind ).c_str(), "w" );
  if( !f )
    { show_error( "Can't open pipe", errno ); return 1; }
  const int wr = std::fwrite( buffer, 1, size, f );
  if( wr != size || pclose( f ) != 0 )
    {
    if( verbosity >= 0 )
      std::fprintf( stderr, "Could not run '%s' : %s.\n",
                    parser.argument( argind ).c_str(), std::strerror( errno ) );
    return 1;
    }

  std::signal( SIGPIPE, SIG_IGN );
  if( verbosity >= 1 ) bits.print();

  const int end = ( ( pos + max_size < size ) ? pos + max_size : size );
  for( int i = pos; i < end; ++i )
    {
    if( verbosity >= 0 )
      std::fprintf( stderr, "byte %d\n", i );
    const uint8_t byte = buffer[i];
    for( int j = 0; j < 255; ++j )
      {
      ++buffer[i];
      if( bits.includes( differing_bits( byte, buffer[i] ) ) )
        {
        f = popen( parser.argument( argind ).c_str(), "w" );
        if( !f )
          { show_error( "Can't open pipe", errno ); return 1; }
        std::fwrite( buffer, 1, size, f );
        if( pclose( f ) == 0 && verbosity >= 0 )
          std::fprintf( stderr, "0x%02X (0x%02X+0x%02X) passed the test\n",
                        buffer[i], byte, j + 1 );
        }
      }
    buffer[i] = byte;
    }

  delete[] buffer;
  return 0;
  }