summaryrefslogtreecommitdiffstats
path: root/decoder.h
blob: 5b06b25da23ea365549b162a05b0c20c45c35bb7 (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
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
/* Lziprecover - Data recovery tool for the lzip format
   Copyright (C) 2009-2022 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/>.
*/

class Range_decoder
  {
  enum { buffer_size = 16384 };
  unsigned long long partial_member_pos;
  uint8_t * const buffer;	// input buffer
  int pos;			// current pos in buffer
  int stream_pos;		// when reached, a new block must be read
  uint32_t code;
  uint32_t range;
  const int infd;		// input file descriptor
  bool at_stream_end;

  bool read_block();

  Range_decoder( const Range_decoder & );	// declared as private
  void operator=( const Range_decoder & );	// declared as private

public:
  explicit Range_decoder( const int ifd )
    :
    partial_member_pos( 0 ),
    buffer( new uint8_t[buffer_size] ),
    pos( 0 ),
    stream_pos( 0 ),
    code( 0 ),
    range( 0xFFFFFFFFU ),
    infd( ifd ),
    at_stream_end( false )
    {}

  ~Range_decoder() { delete[] buffer; }

  unsigned get_code() const { return code; }
  bool finished() { return pos >= stream_pos && !read_block(); }

  unsigned long long member_position() const
    { return partial_member_pos + pos; }

  void reset_member_position()
    { partial_member_pos = 0; partial_member_pos -= pos; }

  uint8_t get_byte()
    {
    // 0xFF avoids decoder error if member is truncated at EOS marker
    if( finished() ) return 0xFF;
    return buffer[pos++];
    }

  int read_data( uint8_t * const outbuf, const int size )
    {
    int sz = 0;
    while( sz < size && !finished() )
      {
      const int rd = std::min( size - sz, stream_pos - pos );
      std::memcpy( outbuf + sz, buffer + pos, rd );
      pos += rd;
      sz += rd;
      }
    return sz;
    }

  /* if ignore_errors, stop reading before the first wrong byte, so that
     unreading the header is not required to sync to next member */
  int read_header_carefully( Lzip_header & header, const bool ignore_errors )
    {
    int sz = 0;
    while( sz < Lzip_header::size && !finished() )
      {
      header.data[sz] = buffer[pos];
      if( ignore_errors &&
          ( ( sz < 4 && header.data[sz] != lzip_magic[sz] ) ||
            ( sz == 4 && !header.verify_version() ) ||
            ( sz == 5 && !isvalid_ds( header.dictionary_size() ) ) ) ) break;
      ++pos; ++sz;
      }
    return sz;
    }

  bool find_header( Lzip_header & header )
    {
    while( !finished() )
      {
      if( buffer[pos] != lzip_magic[0] ) { ++pos; continue; }
      reset_member_position();
      Lzip_header h;
      if( read_header_carefully( h, true ) == Lzip_header::size )
        { header = h; return true; }
      }
    return false;
    }

  void load()
    {
    code = 0;
    for( int i = 0; i < 5; ++i ) code = ( code << 8 ) | get_byte();
    range = 0xFFFFFFFFU;
    code &= range;		// make sure that first byte is discarded
    }

  void normalize()
    {
    if( range <= 0x00FFFFFFU )
      { range <<= 8; code = ( code << 8 ) | get_byte(); }
    }

  unsigned decode( const int num_bits )
    {
    unsigned symbol = 0;
    for( int i = num_bits; i > 0; --i )
      {
      normalize();
      range >>= 1;
//      symbol <<= 1;
//      if( code >= range ) { code -= range; symbol |= 1; }
      const bool bit = ( code >= range );
      symbol <<= 1; symbol += bit;
      code -= range & ( 0U - bit );
      }
    return symbol;
    }

  unsigned decode_bit( Bit_model & bm )
    {
    normalize();
    const uint32_t bound = ( range >> bit_model_total_bits ) * bm.probability;
    if( code < bound )
      {
      range = bound;
      bm.probability +=
        ( bit_model_total - bm.probability ) >> bit_model_move_bits;
      return 0;
      }
    else
      {
      code -= bound;
      range -= bound;
      bm.probability -= bm.probability >> bit_model_move_bits;
      return 1;
      }
    }

  void decode_symbol_bit( Bit_model & bm, unsigned & symbol )
    {
    normalize();
    symbol <<= 1;
    const uint32_t bound = ( range >> bit_model_total_bits ) * bm.probability;
    if( code < bound )
      {
      range = bound;
      bm.probability +=
        ( bit_model_total - bm.probability ) >> bit_model_move_bits;
      }
    else
      {
      code -= bound;
      range -= bound;
      bm.probability -= bm.probability >> bit_model_move_bits;
      symbol |= 1;
      }
    }

  void decode_symbol_bit_reversed( Bit_model & bm, unsigned & model,
                                   unsigned & symbol, const int i )
    {
    normalize();
    model <<= 1;
    const uint32_t bound = ( range >> bit_model_total_bits ) * bm.probability;
    if( code < bound )
      {
      range = bound;
      bm.probability +=
        ( bit_model_total - bm.probability ) >> bit_model_move_bits;
      }
    else
      {
      code -= bound;
      range -= bound;
      bm.probability -= bm.probability >> bit_model_move_bits;
      model |= 1;
      symbol |= 1 << i;
      }
    }

  unsigned decode_tree6( Bit_model bm[] )
    {
    unsigned symbol = 1;
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    return symbol & 0x3F;
    }

  unsigned decode_tree8( Bit_model bm[] )
    {
    unsigned symbol = 1;
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    return symbol & 0xFF;
    }

  unsigned decode_tree_reversed( Bit_model bm[], const int num_bits )
    {
    unsigned model = 1;
    unsigned symbol = 0;
    for( int i = 0; i < num_bits; ++i )
      decode_symbol_bit_reversed( bm[model], model, symbol, i );
    return symbol;
    }

  unsigned decode_tree_reversed4( Bit_model bm[] )
    {
    unsigned model = 1;
    unsigned symbol = 0;
    decode_symbol_bit_reversed( bm[model], model, symbol, 0 );
    decode_symbol_bit_reversed( bm[model], model, symbol, 1 );
    decode_symbol_bit_reversed( bm[model], model, symbol, 2 );
    decode_symbol_bit_reversed( bm[model], model, symbol, 3 );
    return symbol;
    }

  unsigned decode_matched( Bit_model bm[], unsigned match_byte )
    {
    Bit_model * const bm1 = bm + 0x100;
    unsigned symbol = 1;
    while( symbol < 0x100 )
      {
      const unsigned match_bit = ( match_byte <<= 1 ) & 0x100;
      const bool bit = decode_bit( bm1[symbol+match_bit] );
      symbol <<= 1; symbol |= bit;
      if( match_bit >> 8 != bit )
        {
        while( symbol < 0x100 ) decode_symbol_bit( bm[symbol], symbol );
        break;
        }
      }
    return symbol & 0xFF;
    }

  unsigned decode_len( Len_model & lm, const int pos_state )
    {
    Bit_model * bm;
    unsigned mask, offset, symbol = 1;

    if( decode_bit( lm.choice1 ) == 0 )
      { bm = lm.bm_low[pos_state]; mask = 7; offset = 0; goto len3; }
    if( decode_bit( lm.choice2 ) == 0 )
      { bm = lm.bm_mid[pos_state]; mask = 7; offset = len_low_symbols; goto len3; }
    bm = lm.bm_high; mask = 0xFF; offset = len_low_symbols + len_mid_symbols;
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
len3:
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    decode_symbol_bit( bm[symbol], symbol );
    return ( symbol & mask ) + min_match_len + offset;
    }
  };


class LZ_decoder
  {
  const unsigned long long outskip;
  const unsigned long long outend;
  unsigned long long partial_data_pos;
  Range_decoder & rdec;
  const unsigned dictionary_size;
  uint8_t * const buffer;	// output buffer
  unsigned pos;			// current pos in buffer
  unsigned stream_pos;		// first byte not yet written to file
  uint32_t crc_;
  const int outfd;		// output file descriptor
  bool pos_wrapped;

  unsigned long long stream_position() const
    { return partial_data_pos + stream_pos; }
  void flush_data();
  bool verify_trailer( const Pretty_print & pp ) const;

  uint8_t peek_prev() const
    { return buffer[((pos > 0) ? pos : dictionary_size)-1]; }

  uint8_t peek( const unsigned distance ) const
    {
    const unsigned i = ( ( pos > distance ) ? 0 : dictionary_size ) +
                       pos - distance - 1;
    return buffer[i];
    }

  void put_byte( const uint8_t b )
    {
    buffer[pos] = b;
    if( ++pos >= dictionary_size ) flush_data();
    }

  void copy_block( const unsigned distance, unsigned len )
    {
    unsigned lpos = pos, i = lpos - distance - 1;
    bool fast, fast2;
    if( lpos > distance )
      {
      fast = ( len < dictionary_size - lpos );
      fast2 = ( fast && len <= lpos - i );
      }
    else
      {
      i += dictionary_size;
      fast = ( len < dictionary_size - i );	// (i == pos) may happen
      fast2 = ( fast && len <= i - lpos );
      }
    if( fast )					// no wrap
      {
      pos += len;
      if( fast2 )				// no wrap, no overlap
        std::memcpy( buffer + lpos, buffer + i, len );
      else
        for( ; len > 0; --len ) buffer[lpos++] = buffer[i++];
      }
    else for( ; len > 0; --len )
      {
      buffer[pos] = buffer[i];
      if( ++pos >= dictionary_size ) flush_data();
      if( ++i >= dictionary_size ) i = 0;
      }
    }

  LZ_decoder( const LZ_decoder & );		// declared as private
  void operator=( const LZ_decoder & );		// declared as private

public:
  LZ_decoder( Range_decoder & rde, const unsigned dict_size, const int ofd,
              const unsigned long long oskip = 0,
              const unsigned long long oend = -1ULL )
    :
    outskip( oskip ),
    outend( oend ),
    partial_data_pos( 0 ),
    rdec( rde ),
    dictionary_size( dict_size ),
    buffer( new uint8_t[dictionary_size] ),
    pos( 0 ),
    stream_pos( 0 ),
    crc_( 0xFFFFFFFFU ),
    outfd( ofd ),
    pos_wrapped( false )
    // prev_byte of first byte; also for peek( 0 ) on corrupt file
    { buffer[dictionary_size-1] = 0; }

  ~LZ_decoder() { delete[] buffer; }

  unsigned crc() const { return crc_ ^ 0xFFFFFFFFU; }
  unsigned long long data_position() const { return partial_data_pos + pos; }

  int decode_member( const Pretty_print & pp );
  };