/*  Lziprecover - Data recovery tool for the lzip format
    Copyright (C) 2009-2014 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/>.
*/

#define _FILE_OFFSET_BITS 64

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

#include "lzip.h"
#include "mtester.h"


void LZ_mtester::flush_data()
  {
  if( pos > stream_pos )
    {
    const int size = pos - stream_pos;
    crc32.update_buf( crc_, buffer + stream_pos, size );
    if( pos >= buffer_size ) { partial_data_pos += pos; pos = 0; }
    stream_pos = pos;
    }
  }


bool LZ_mtester::verify_trailer()
  {
  const File_trailer * trailer = rdec.get_trailer();
  if( !trailer ) return false;

  return ( rdec.code_is_zero() &&
           trailer->data_crc() == crc() &&
           trailer->data_size() == data_position() &&
           trailer->member_size() == (unsigned long)member_position() );
  }


void LZ_mtester::duplicate_buffer()
  {
  uint8_t * const tmp = new uint8_t[buffer_size];
  if( data_position() > 0 )
    std::memcpy( tmp, buffer, std::min( data_position(),
                                        (unsigned long long)buffer_size ) );
  else tmp[buffer_size-1] = 0;		// prev_byte of first byte
  buffer = tmp;
  }


/* Return value: 0 = OK, 1 = decoder error, 2 = unexpected EOF,
                 3 = trailer error, 4 = unknown marker found,
                 -1 = pos_limit reached. */
int LZ_mtester::test_member( const long pos_limit )
  {
  if( pos_limit < File_header::size + 5 ) return -1;
  if( member_position() == File_header::size ) rdec.load();
  while( !rdec.finished() )
    {
    if( member_position() >= pos_limit ) { flush_data(); return -1; }
    const int pos_state = data_position() & pos_state_mask;
    if( rdec.decode_bit( bm_match[state()][pos_state] ) == 0 )	// 1st bit
      {
      const uint8_t prev_byte = get_prev_byte();
      if( state.is_char() )
        {
        state.set_char1();
        put_byte( rdec.decode_tree8( bm_literal[get_lit_state(prev_byte)] ) );
        }
      else
        {
        state.set_char2();
        put_byte( rdec.decode_matched( bm_literal[get_lit_state(prev_byte)],
                                       get_byte( rep0 ) ) );
        }
      }
    else
      {
      int len;
      if( rdec.decode_bit( bm_rep[state()] ) != 0 )		// 2nd bit
        {
        if( rdec.decode_bit( bm_rep0[state()] ) != 0 )		// 3rd bit
          {
          unsigned distance;
          if( rdec.decode_bit( bm_rep1[state()] ) == 0 )	// 4th bit
            distance = rep1;
          else
            {
            if( rdec.decode_bit( bm_rep2[state()] ) == 0 )	// 5th bit
              distance = rep2;
            else
              { distance = rep3; rep3 = rep2; }
            rep2 = rep1;
            }
          rep1 = rep0;
          rep0 = distance;
          }
        else
          {
          if( rdec.decode_bit( bm_len[state()][pos_state] ) == 0 ) // 4th bit
            { state.set_short_rep(); put_byte( get_byte( rep0 ) ); continue; }
          }
        state.set_rep();
        len = min_match_len + rdec.decode_len( rep_len_model, pos_state );
        }
      else
        {
        const unsigned rep0_saved = rep0;
        len = min_match_len + rdec.decode_len( match_len_model, pos_state );
        const int dis_slot = rdec.decode_tree6( bm_dis_slot[get_len_state(len)] );
        if( dis_slot < start_dis_model ) rep0 = dis_slot;
        else
          {
          const int direct_bits = ( dis_slot >> 1 ) - 1;
          rep0 = ( 2 | ( dis_slot & 1 ) ) << direct_bits;
          if( dis_slot < end_dis_model )
            rep0 += rdec.decode_tree_reversed( bm_dis + rep0 - dis_slot - 1,
                                               direct_bits );
          else
            {
            rep0 += rdec.decode( direct_bits - dis_align_bits ) << dis_align_bits;
            rep0 += rdec.decode_tree_reversed4( bm_align );
            if( rep0 == 0xFFFFFFFFU )		// marker found
              {
              rep0 = rep0_saved;
              rdec.normalize();
              flush_data();
              if( len == min_match_len )	// End Of Stream marker
                {
                if( verify_trailer() ) return 0; else return 3;
                }
              return 4;
              }
            }
          }
        rep3 = rep2; rep2 = rep1; rep1 = rep0_saved;
        state.set_match();
        if( rep0 >= dictionary_size || rep0 >= data_position() )
          { flush_data(); return 1; }
        }
      copy_block( rep0, len );
      }
    }
  flush_data();
  return 2;
  }


uint8_t * read_member( const int infd, const long long mpos,
                       const long long msize )
  {
  if( msize <= 0 || msize > LONG_MAX )
    { show_error( "Member is larger than LONG_MAX." ); return 0; }
  if( !safe_seek( infd, mpos ) ) return 0;
  uint8_t * const buffer = new uint8_t[msize];

  if( readblock( infd, buffer, msize ) != msize )
    { show_error( "Error reading input file", errno );
      delete[] buffer; return 0; }
  return buffer;
  }


const LZ_mtester * prepare_master( const uint8_t * const buffer,
                                   const long buffer_size,
                                   const long pos_limit )
  {
  File_header & header = *(File_header *)buffer;
  const unsigned dictionary_size = header.dictionary_size();
  if( header.verify_magic() && header.verify_version() &&
      dictionary_size >= min_dictionary_size &&
      dictionary_size <= max_dictionary_size )
    {
    LZ_mtester * const master =
      new LZ_mtester( buffer, buffer_size, dictionary_size );
    if( master->test_member( pos_limit ) == -1 ) return master;
    delete master;
    }
  return 0;
  }


bool test_member_rest( const LZ_mtester & master, long * const failure_posp )
  {
  LZ_mtester mtester( master );
  mtester.duplicate_buffer();
  if( mtester.test_member() == 0 && mtester.finished() ) return true;
  if( failure_posp ) *failure_posp = mtester.member_position();
  return false;
  }