summaryrefslogtreecommitdiffstats
path: root/file_index.cc
blob: 581b516c8d82e2c0a48309340f60207ae612f6fc (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
/*  Plzip - Parallel compressor compatible with lzip
    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
    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 <cstdio>
#include <cstring>
#include <string>
#include <vector>
#include <pthread.h>
#include <stdint.h>
#include <unistd.h>

#include "lzip.h"
#include "file_index.h"


namespace {

int seek_read( const int fd, uint8_t * const buf, const int size,
               const long long pos )
  {
  if( lseek( fd, pos, SEEK_SET ) == pos )
    return readblock( fd, buf, size );
  return 0;
  }

} // end namespace


void File_index::set_errno_error( const char * const msg )
  {
  error_ = msg; error_ += std::strerror( errno );
  retval_ = 1;
  }

void File_index::set_num_error( const char * const msg, unsigned long long num )
  {
  char buf[80];
  snprintf( buf, sizeof buf, "%s%llu", msg, num );
  error_ = buf;
  retval_ = 2;
  }


// If successful, push last member and set pos to member header.
bool File_index::skip_trailing_data( const int fd, long long & pos )
  {
  enum { block_size = 16384,
         buffer_size = block_size + File_trailer::size - 1 + File_header::size };
  uint8_t buffer[buffer_size];
  if( pos < min_member_size ) return false;
  int bsize = pos % block_size;			// total bytes in buffer
  if( bsize <= buffer_size - block_size ) bsize += block_size;
  int search_size = bsize;			// bytes to search for trailer
  int rd_size = bsize;				// bytes to read from file
  unsigned long long ipos = pos - rd_size;	// aligned to block_size

  while( true )
    {
    if( seek_read( fd, buffer, rd_size, ipos ) != rd_size )
      { set_errno_error( "Error seeking member trailer: " ); return false; }
    const uint8_t max_msb = ( ipos + search_size ) >> 56;
    for( int i = search_size; i >= File_trailer::size; --i )
      if( buffer[i-1] <= max_msb )	// most significant byte of member_size
        {
        File_trailer & trailer =
          *(File_trailer *)( buffer + i - File_trailer::size );
        const unsigned long long member_size = trailer.member_size();
        if( member_size == 0 )
          { while( i > File_trailer::size && buffer[i-9] == 0 ) --i; continue; }
        if( member_size < min_member_size || member_size > ipos + i )
          continue;
        File_header header;
        if( seek_read( fd, header.data, File_header::size,
                       ipos + i - member_size ) != File_header::size )
          { set_errno_error( "Error reading member header: " ); return false; }
        const unsigned dictionary_size = header.dictionary_size();
        if( !header.verify_magic() || !header.verify_version() ||
            !isvalid_ds( dictionary_size ) ) continue;
        if( (*(File_header *)( buffer + i )).verify_prefix( bsize - i ) )
          {
          error_ = "Last member in input file is truncated or corrupt.";
          retval_ = 2; return false;
          }
        pos = ipos + i - member_size;
        member_vector.push_back( Member( 0, trailer.data_size(), pos,
                                         member_size, dictionary_size ) );
        return true;
        }
    if( ipos <= 0 )
      { set_num_error( "Member size in trailer is corrupt at pos ", pos - 8 );
        return false; }
    bsize = buffer_size;
    search_size = bsize - File_header::size;
    rd_size = block_size;
    ipos -= rd_size;
    std::memcpy( buffer + rd_size, buffer, buffer_size - rd_size );
    }
  }


File_index::File_index( const int infd, const bool ignore_trailing )
  : isize( lseek( infd, 0, SEEK_END ) ), retval_( 0 )
  {
  if( isize < 0 )
    { set_errno_error( "Input file is not seekable: " ); return; }
  if( isize < min_member_size )
    { error_ = "Input file is too short."; retval_ = 2; return; }
  if( isize > INT64_MAX )
    { error_ = "Input file is too long (2^63 bytes or more).";
      retval_ = 2; return; }

  File_header header;
  if( seek_read( infd, header.data, File_header::size, 0 ) != File_header::size )
    { set_errno_error( "Error reading member header: " ); return; }
  if( !header.verify_magic() )
    { error_ = bad_magic_msg; retval_ = 2; return; }
  if( !header.verify_version() )
    { error_ = bad_version( header.version() ); retval_ = 2; return; }
  if( !isvalid_ds( header.dictionary_size() ) )
    { error_ = bad_dict_msg; retval_ = 2; return; }

  long long pos = isize;	// always points to a header or to EOF
  while( pos >= min_member_size )
    {
    File_trailer trailer;
    if( seek_read( infd, trailer.data, File_trailer::size,
                   pos - File_trailer::size ) != File_trailer::size )
      { set_errno_error( "Error reading member trailer: " ); break; }
    const unsigned long long member_size = trailer.member_size();
    if( member_size < min_member_size || member_size > (unsigned long long)pos )
      {
      if( !member_vector.empty() )
        set_num_error( "Member size in trailer is corrupt at pos ", pos - 8 );
      else if( skip_trailing_data( infd, pos ) )
        { if( ignore_trailing ) continue;
          error_ = trailing_msg; retval_ = 2; return; }
      break;
      }
    if( seek_read( infd, header.data, File_header::size,
                   pos - member_size ) != File_header::size )
      { set_errno_error( "Error reading member header: " ); break; }
    const unsigned dictionary_size = header.dictionary_size();
    if( !header.verify_magic() || !header.verify_version() ||
        !isvalid_ds( dictionary_size ) )
      {
      if( !member_vector.empty() )
        set_num_error( "Bad header at pos ", pos - member_size );
      else if( skip_trailing_data( infd, pos ) )
        { if( ignore_trailing ) continue;
          error_ = trailing_msg; retval_ = 2; return; }
      break;
      }
    pos -= member_size;
    member_vector.push_back( Member( 0, trailer.data_size(), pos,
                                     member_size, dictionary_size ) );
    }
  if( pos != 0 || member_vector.empty() )
    {
    member_vector.clear();
    if( retval_ == 0 ) { error_ = "Can't create file index."; retval_ = 2; }
    return;
    }
  std::reverse( member_vector.begin(), member_vector.end() );
  for( unsigned long i = 0; i < member_vector.size() - 1; ++i )
    {
    const long long end = member_vector[i].dblock.end();
    if( end < 0 || end > INT64_MAX )
      {
      member_vector.clear();
      error_ = "Data in input file is too long (2^63 bytes or more).";
      retval_ = 2; return;
      }
    member_vector[i+1].dblock.pos( end );
    }
  }