summaryrefslogtreecommitdiffstats
path: root/encoder.c
blob: 04080631b6cecf54619ee6242419f4f2f365f353 (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
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
/*  Lzlib - Compression library for lzip files
    Copyright (C) 2009, 2010, 2011, 2012, 2013 Antonio Diaz Diaz.

    This library 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 3 of the License, or
    (at your option) any later version.

    This library 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 library.  If not, see <http://www.gnu.org/licenses/>.

    As a special exception, you may use this file as part of a free
    software library without restriction.  Specifically, if other files
    instantiate templates or use macros or inline functions from this
    file, or you compile this file and link it with other files to
    produce an executable, this file does not by itself cause the
    resulting executable to be covered by the GNU General Public
    License.  This exception does not however invalidate any other
    reasons why the executable file might be covered by the GNU General
    Public License.
*/

static bool Mf_normalize_pos( struct Matchfinder * const mf )
  {
  if( mf->pos > mf->stream_pos )
    { mf->pos = mf->stream_pos; return false; }
  if( !mf->at_stream_end )
    {
    int i;
    const int offset = mf->pos - mf->dictionary_size - before_size;
    const int size = mf->stream_pos - offset;
    memmove( mf->buffer, mf->buffer + offset, size );
    mf->partial_data_pos += offset;
    mf->pos -= offset;
    mf->stream_pos -= offset;
    for( i = 0; i < mf->num_prev_positions; ++i )
      if( mf->prev_positions[i] >= 0 ) mf->prev_positions[i] -= offset;
    for( i = 0; i < 2 * ( mf->dictionary_size + 1 ); ++i )
      if( mf->prev_pos_tree[i] >= 0 ) mf->prev_pos_tree[i] -= offset;
    }
  return true;
  }


static bool Mf_init( struct Matchfinder * const mf,
                     const int dict_size, const int match_len_limit )
  {
  const int buffer_size_limit = ( 2 * dict_size ) + before_size + after_size;
  unsigned size;
  int i;

  mf->partial_data_pos = 0;
  mf->match_len_limit = match_len_limit;
  mf->pos = 0;
  mf->cyclic_pos = 0;
  mf->stream_pos = 0;
  mf->cycles = ( match_len_limit < max_match_len ) ?
               16 + ( match_len_limit / 2 ) : 256;
  mf->at_stream_end = false;
  mf->been_flushed = false;
  mf->flushing = false;

  mf->buffer_size = max( 65536, buffer_size_limit );
  mf->buffer = (uint8_t *)malloc( mf->buffer_size );
  if( !mf->buffer ) return false;
  mf->dictionary_size = dict_size;
  mf->pos_limit = mf->buffer_size - after_size;
  size = 1 << max( 16, real_bits( mf->dictionary_size - 1 ) - 2 );
  if( mf->dictionary_size > 1 << 26 )		/* 64 MiB */
    size >>= 1;
  mf->key4_mask = size - 1;
  size += num_prev_positions2;
  size += num_prev_positions3;

  mf->num_prev_positions = size;
  size += ( 2 * ( mf->dictionary_size + 1 ) );
  if( size * sizeof (int32_t) <= size ) mf->prev_positions = 0;
  else mf->prev_positions = (int32_t *)malloc( size * sizeof (int32_t) );
  if( !mf->prev_positions ) { free( mf->buffer ); return false; }
  mf->prev_pos_tree = mf->prev_positions + mf->num_prev_positions;
  for( i = 0; i < mf->num_prev_positions; ++i ) mf->prev_positions[i] = -1;
  return true;
  }


static void Mf_adjust_distionary_size( struct Matchfinder * const mf )
  {
  if( mf->stream_pos < mf->dictionary_size )
    {
    int size;
    mf->buffer_size =
    mf->dictionary_size =
    mf->pos_limit = max( min_dictionary_size, mf->stream_pos );
    size = 1 << max( 16, real_bits( mf->dictionary_size - 1 ) - 2 );
    if( mf->dictionary_size > 1 << 26 )
      size >>= 1;
    mf->key4_mask = size - 1;
    size += num_prev_positions2;
    size += num_prev_positions3;
    mf->num_prev_positions = size;
    mf->prev_pos_tree = mf->prev_positions + mf->num_prev_positions;
    }
  }


static void Mf_reset( struct Matchfinder * const mf )
  {
  int i;
  const int size = mf->stream_pos - mf->pos;
  if( size > 0 ) memmove( mf->buffer, mf->buffer + mf->pos, size );
  mf->partial_data_pos = 0;
  mf->stream_pos -= mf->pos;
  mf->pos = 0;
  mf->cyclic_pos = 0;
  mf->at_stream_end = false;
  mf->been_flushed = false;
  mf->flushing = false;
  for( i = 0; i < mf->num_prev_positions; ++i ) mf->prev_positions[i] = -1;
  }


static int Mf_get_match_pairs( struct Matchfinder * const mf, struct Pair * pairs )
  {
  int32_t * ptr0 = mf->prev_pos_tree + ( mf->cyclic_pos << 1 );
  int32_t * ptr1 = ptr0 + 1;
  int32_t * newptr;
  int len = 0, len0 = 0, len1 = 0;
  int maxlen = min_match_len - 1;
  int num_pairs = 0;
  const int min_pos = (mf->pos > mf->dictionary_size) ?
                       mf->pos - mf->dictionary_size : 0;
  const uint8_t * const data = mf->buffer + mf->pos;
  int count, delta, key2, key3, key4, newpos;
  unsigned tmp;
  int len_limit = mf->match_len_limit;

  if( len_limit > Mf_available_bytes( mf ) )
    {
    mf->been_flushed = true;
    len_limit = Mf_available_bytes( mf );
    if( len_limit < 4 ) { *ptr0 = *ptr1 = -1; return 0; }
    }

  tmp = crc32[data[0]] ^ data[1];
  key2 = tmp & ( num_prev_positions2 - 1 );
  tmp ^= (uint32_t)data[2] << 8;
  key3 = num_prev_positions2 + ( tmp & ( num_prev_positions3 - 1 ) );
  key4 = num_prev_positions2 + num_prev_positions3 +
         ( ( tmp ^ ( crc32[data[3]] << 5 ) ) & mf->key4_mask );

  if( pairs )
    {
    int np2 = mf->prev_positions[key2];
    int np3 = mf->prev_positions[key3];
    if( np2 >= min_pos && mf->buffer[np2] == data[0] )
      {
      pairs[0].dis = mf->pos - np2 - 1;
      pairs[0].len = maxlen = 2;
      num_pairs = 1;
      }
    if( np2 != np3 && np3 >= min_pos && mf->buffer[np3] == data[0] )
      {
      maxlen = 3;
      pairs[num_pairs].dis = mf->pos - np3 - 1;
      ++num_pairs;
      np2 = np3;
      }
    if( num_pairs > 0 )
      {
      delta = mf->pos - np2;
      while( maxlen < len_limit && data[maxlen-delta] == data[maxlen] )
        ++maxlen;
      pairs[num_pairs-1].len = maxlen;
      if( maxlen >= len_limit ) pairs = 0;
      }
    if( maxlen < 3 ) maxlen = 3;
    }

  mf->prev_positions[key2] = mf->pos;
  mf->prev_positions[key3] = mf->pos;
  newpos = mf->prev_positions[key4];
  mf->prev_positions[key4] = mf->pos;

  for( count = mf->cycles; ; )
    {
    if( newpos < min_pos || --count < 0 ) { *ptr0 = *ptr1 = -1; break; }

    if( mf->been_flushed ) len = 0;
    delta = mf->pos - newpos;
    newptr = mf->prev_pos_tree +
      ( ( mf->cyclic_pos - delta +
          ( (mf->cyclic_pos >= delta) ? 0 : mf->dictionary_size + 1 ) ) << 1 );
    if( data[len-delta] == data[len] )
      {
      while( ++len < len_limit && data[len-delta] == data[len] ) {}
      if( pairs && maxlen < len )
        {
        pairs[num_pairs].dis = delta - 1;
        pairs[num_pairs].len = maxlen = len;
        ++num_pairs;
        }
      if( len >= len_limit )
        {
        *ptr0 = newptr[0];
        *ptr1 = newptr[1];
        break;
        }
      }
    if( data[len-delta] < data[len] )
      {
      *ptr0 = newpos;
      ptr0 = newptr + 1;
      newpos = *ptr0;
      len0 = len; if( len1 < len ) len = len1;
      }
    else
      {
      *ptr1 = newpos;
      ptr1 = newptr;
      newpos = *ptr1;
      len1 = len; if( len0 < len ) len = len0;
      }
    }
  return num_pairs;
  }


static void Lee_encode( struct Len_encoder * const len_encoder,
                        struct Range_encoder * const renc,
                        int symbol, const int pos_state )
  {
  symbol -= min_match_len;
  if( symbol < len_low_symbols )
    {
    Re_encode_bit( renc, &len_encoder->lm.choice1, 0 );
    Re_encode_tree( renc, len_encoder->lm.bm_low[pos_state], symbol, len_low_bits );
    }
  else
    {
    Re_encode_bit( renc, &len_encoder->lm.choice1, 1 );
    if( symbol < len_low_symbols + len_mid_symbols )
      {
      Re_encode_bit( renc, &len_encoder->lm.choice2, 0 );
      Re_encode_tree( renc, len_encoder->lm.bm_mid[pos_state],
                      symbol - len_low_symbols, len_mid_bits );
      }
    else
      {
      Re_encode_bit( renc, &len_encoder->lm.choice2, 1 );
      Re_encode_tree( renc, len_encoder->lm.bm_high,
                      symbol - len_low_symbols - len_mid_symbols, len_high_bits );
      }
    }
  if( --len_encoder->counters[pos_state] <= 0 )
    Lee_update_prices( len_encoder, pos_state );
  }


     /* End Of Stream mark => (dis == 0xFFFFFFFFU, len == min_match_len) */
static bool LZe_full_flush( struct LZ_encoder * const encoder, const State state )
  {
  int i;
  const int pos_state = Mf_data_position( encoder->matchfinder ) & pos_state_mask;
  File_trailer trailer;
  if( encoder->member_finished ||
      Cb_free_bytes( &encoder->renc.cb ) < max_marker_size + Ft_size )
    return false;
  Re_encode_bit( &encoder->renc, &encoder->bm_match[state][pos_state], 1 );
  Re_encode_bit( &encoder->renc, &encoder->bm_rep[state], 0 );
  LZe_encode_pair( encoder, 0xFFFFFFFFU, min_match_len, pos_state );
  Re_flush( &encoder->renc );
  Ft_set_data_crc( trailer, LZe_crc( encoder ) );
  Ft_set_data_size( trailer, Mf_data_position( encoder->matchfinder ) );
  Ft_set_member_size( trailer, Re_member_position( &encoder->renc ) + Ft_size );
  for( i = 0; i < Ft_size; ++i )
    Cb_put_byte( &encoder->renc.cb, trailer[i] );
  return true;
  }


     /* Sync Flush mark => (dis == 0xFFFFFFFFU, len == min_match_len + 1) */
static bool LZe_sync_flush( struct LZ_encoder * const encoder )
  {
  const int pos_state = Mf_data_position( encoder->matchfinder ) & pos_state_mask;
  const State state = encoder->state;
  if( encoder->member_finished ||
      Cb_free_bytes( &encoder->renc.cb ) < max_marker_size )
    return false;
  Re_encode_bit( &encoder->renc, &encoder->bm_match[state][pos_state], 1 );
  Re_encode_bit( &encoder->renc, &encoder->bm_rep[state], 0 );
  LZe_encode_pair( encoder, 0xFFFFFFFFU, min_match_len + 1, pos_state );
  Re_flush( &encoder->renc );
  return true;
  }


static void LZe_fill_align_prices( struct LZ_encoder * const encoder )
  {
  int i;
  for( i = 0; i < dis_align_size; ++i )
    encoder->align_prices[i] =
      price_symbol_reversed( encoder->bm_align, i, dis_align_bits );
  encoder->align_price_count = dis_align_size;
  }


static void LZe_fill_distance_prices( struct LZ_encoder * const encoder )
  {
  int dis, dis_state;
  for( dis = start_dis_model; dis < modeled_distances; ++dis )
    {
    const int dis_slot = dis_slots[dis];
    const int direct_bits = ( dis_slot >> 1 ) - 1;
    const int base = ( 2 | ( dis_slot & 1 ) ) << direct_bits;
    const int price =
      price_symbol_reversed( encoder->bm_dis + base - dis_slot - 1,
                             dis - base, direct_bits );
    for( dis_state = 0; dis_state < max_dis_states; ++dis_state )
      encoder->dis_prices[dis_state][dis] = price;
    }

  for( dis_state = 0; dis_state < max_dis_states; ++dis_state )
    {
    int * const dsp = encoder->dis_slot_prices[dis_state];
    int * const dp = encoder->dis_prices[dis_state];
    const Bit_model * const bmds = encoder->bm_dis_slot[dis_state];
    int slot = 0;
    for( ; slot < end_dis_model && slot < encoder->num_dis_slots; ++slot )
      dsp[slot] = price_symbol( bmds, slot, dis_slot_bits );
    for( ; slot < encoder->num_dis_slots; ++slot )
      dsp[slot] = price_symbol( bmds, slot, dis_slot_bits ) +
                  (((( slot >> 1 ) - 1 ) - dis_align_bits ) << price_shift_bits );

    for( dis = 0; dis < start_dis_model; ++dis )
      dp[dis] = dsp[dis];
    for( ; dis < modeled_distances; ++dis )
      dp[dis] += dsp[dis_slots[dis]];
    }
  }


static bool LZe_init( struct LZ_encoder * const encoder,
                      struct Matchfinder * const mf,
                      const File_header header,
                      const unsigned long long member_size )
  {
  int i;
  encoder->member_size_limit = member_size - Ft_size - max_marker_size;
  encoder->pending_num_pairs = 0;
  encoder->crc = 0xFFFFFFFFU;

  Bm_array_init( encoder->bm_literal[0], (1 << literal_context_bits) * 0x300 );
  Bm_array_init( encoder->bm_match[0], states * pos_states );
  Bm_array_init( encoder->bm_rep, states );
  Bm_array_init( encoder->bm_rep0, states );
  Bm_array_init( encoder->bm_rep1, states );
  Bm_array_init( encoder->bm_rep2, states );
  Bm_array_init( encoder->bm_len[0], states * pos_states );
  Bm_array_init( encoder->bm_dis_slot[0], max_dis_states * (1 << dis_slot_bits) );
  Bm_array_init( encoder->bm_dis, modeled_distances - end_dis_model );
  Bm_array_init( encoder->bm_align, dis_align_size );

  encoder->matchfinder = mf;
  if( !Re_init( &encoder->renc ) ) return false;
  Lee_init( &encoder->match_len_encoder, encoder->matchfinder->match_len_limit );
  Lee_init( &encoder->rep_len_encoder, encoder->matchfinder->match_len_limit );
  encoder->num_dis_slots =
    2 * real_bits( encoder->matchfinder->dictionary_size - 1 );

  for( i = 0; i < num_rep_distances; ++i ) encoder->rep_distances[i] = 0;
  encoder->align_price_count = 0;
  encoder->fill_counter = 0;
  encoder->state = 0;
  encoder->member_finished = false;

  for( i = 0; i < Fh_size; ++i )
    Cb_put_byte( &encoder->renc.cb, header[i] );
  return true;
  }


/* Return value == number of bytes advanced (ahead).
   trials[0]..trials[ahead-1] contain the steps to encode.
   ( trials[0].dis == -1 && trials[0].price == 1 ) means literal.
*/
static int LZe_sequence_optimizer( struct LZ_encoder * const encoder,
                                   const int reps[num_rep_distances],
                                   const State state )
  {
  int main_len, num_pairs, i, rep, cur = 0, num_trials, len;
  int replens[num_rep_distances];
  int rep_index = 0;

  if( encoder->pending_num_pairs > 0 )		/* from previous call */
    {
    num_pairs = encoder->pending_num_pairs;
    encoder->pending_num_pairs = 0;
    }
  else
    num_pairs = LZe_read_match_distances( encoder );
  main_len = ( num_pairs > 0 ) ? encoder->pairs[num_pairs-1].len : 0;

  for( i = 0; i < num_rep_distances; ++i )
    {
    replens[i] =
      Mf_true_match_len( encoder->matchfinder, 0, reps[i] + 1, max_match_len );
    if( replens[i] > replens[rep_index] ) rep_index = i;
    }
  if( replens[rep_index] >= encoder->matchfinder->match_len_limit )
    {
    encoder->trials[0].dis = rep_index;
    encoder->trials[0].price = replens[rep_index];
    if( !LZe_move_pos( encoder, replens[rep_index] ) ) return 0;
    return replens[rep_index];
    }

  if( main_len >= encoder->matchfinder->match_len_limit )
    {
    encoder->trials[0].dis = encoder->pairs[num_pairs-1].dis + num_rep_distances;
    encoder->trials[0].price = main_len;
    if( !LZe_move_pos( encoder, main_len ) ) return 0;
    return main_len;
    }

  {
  const int pos_state = Mf_data_position( encoder->matchfinder ) & pos_state_mask;
  const int match_price = price1( encoder->bm_match[state][pos_state] );
  const int rep_match_price = match_price + price1( encoder->bm_rep[state] );
  const uint8_t prev_byte = Mf_peek( encoder->matchfinder, -1 );
  const uint8_t cur_byte = Mf_peek( encoder->matchfinder, 0 );
  const uint8_t match_byte = Mf_peek( encoder->matchfinder, -reps[0]-1 );

  encoder->trials[0].state = state;
  encoder->trials[1].dis = -1;
  encoder->trials[1].price = price0( encoder->bm_match[state][pos_state] );
  if( St_is_char( state ) )
    encoder->trials[1].price +=
      LZe_price_literal( encoder, prev_byte, cur_byte );
  else
    encoder->trials[1].price +=
      LZe_price_matched( encoder, prev_byte, cur_byte, match_byte );

  if( match_byte == cur_byte )
    Tr_update( &encoder->trials[1], rep_match_price +
               LZe_price_rep_len1( encoder, state, pos_state ), 0, 0 );

  num_trials = max( main_len, replens[rep_index] );

  if( num_trials < min_match_len )
    {
    encoder->trials[0].dis = encoder->trials[1].dis;
    encoder->trials[0].price = 1;
    if( !Mf_move_pos( encoder->matchfinder ) ) return 0;
    return 1;
    }

  for( i = 0; i < num_rep_distances; ++i )
    encoder->trials[0].reps[i] = reps[i];
  encoder->trials[1].prev_index = 0;
  encoder->trials[1].prev_index2 = single_step_trial;

  for( len = min_match_len; len <= num_trials; ++len )
    encoder->trials[len].price = infinite_price;

  for( rep = 0; rep < num_rep_distances; ++rep )
    {
    int price;
    if( replens[rep] < min_match_len ) continue;
    price = rep_match_price + LZe_price_rep( encoder, rep, state, pos_state );

    for( len = min_match_len; len <= replens[rep]; ++len )
      Tr_update( &encoder->trials[len], price +
                 Lee_price( &encoder->rep_len_encoder, len, pos_state ),
                 rep, 0 );
    }

  if( main_len > replens[0] )
    {
    const int normal_match_price = match_price + price0( encoder->bm_rep[state] );
    i = 0, len = max( replens[0] + 1, min_match_len );
    while( len > encoder->pairs[i].len ) ++i;
    while( true )
      {
      const int dis = encoder->pairs[i].dis;
      Tr_update( &encoder->trials[len], normal_match_price +
                 LZe_price_pair( encoder, dis, len, pos_state ),
                 dis + num_rep_distances, 0 );
      if( ++len > encoder->pairs[i].len && ++i >= num_pairs ) break;
      }
    }
  }

  if( !Mf_move_pos( encoder->matchfinder ) ) return 0;

  while( true )				/* price optimization loop */
    {
    struct Trial *cur_trial, *next_trial;
    int newlen, pos_state, prev_index, prev_index2, available_bytes, len_limit;
    int start_len = min_match_len;
    int next_price, match_price, rep_match_price;
    State cur_state;
    uint8_t prev_byte, cur_byte, match_byte;

    if( ++cur >= num_trials )		/* no more initialized trials */
      {
      LZe_backward( encoder, cur );
      return cur;
      }

    num_pairs = LZe_read_match_distances( encoder );
    newlen = ( num_pairs > 0 ) ? encoder->pairs[num_pairs-1].len : 0;
    if( newlen >= encoder->matchfinder->match_len_limit )
      {
      encoder->pending_num_pairs = num_pairs;
      LZe_backward( encoder, cur );
      return cur;
      }

    /* give final values to current trial */
    cur_trial = &encoder->trials[cur];
    prev_index = cur_trial->prev_index;
    prev_index2 = cur_trial->prev_index2;

    if( prev_index2 != single_step_trial )
      {
      --prev_index;
      if( prev_index2 >= 0 )
        {
        cur_state = encoder->trials[prev_index2].state;
        if( cur_trial->dis2 < num_rep_distances )
          cur_state = St_set_rep( cur_state );
        else
          cur_state = St_set_match( cur_state );
        }
      else
        cur_state = encoder->trials[prev_index].state;
      cur_state = St_set_char( cur_state );
      }
    else
      cur_state = encoder->trials[prev_index].state;

    if( prev_index == cur - 1 )
      {
      if( cur_trial->dis == 0 )
        cur_state = St_set_short_rep( cur_state );
      else
        cur_state = St_set_char( cur_state );
      for( i = 0; i < num_rep_distances; ++i )
        cur_trial->reps[i] = encoder->trials[prev_index].reps[i];
      }
    else
      {
      int dis;
      if( prev_index2 >= 0 )
        {
        dis = cur_trial->dis2;
        prev_index = prev_index2;
        cur_state = St_set_rep( cur_state );
        }
      else
        {
        dis = cur_trial->dis;
        if( dis < num_rep_distances )
          cur_state = St_set_rep( cur_state );
        else
          cur_state = St_set_match( cur_state );
        }
      for( i = 0; i < num_rep_distances; ++i )
        cur_trial->reps[i] = encoder->trials[prev_index].reps[i];
      LZe_mtf_reps( dis, cur_trial->reps );
      }
    cur_trial->state = cur_state;

    pos_state = Mf_data_position( encoder->matchfinder ) & pos_state_mask;
    prev_byte = Mf_peek( encoder->matchfinder, -1 );
    cur_byte = Mf_peek( encoder->matchfinder, 0 );
    match_byte = Mf_peek( encoder->matchfinder, -cur_trial->reps[0]-1 );

    next_price = cur_trial->price +
                 price0( encoder->bm_match[cur_state][pos_state] );
    if( St_is_char( cur_state ) )
      next_price += LZe_price_literal( encoder, prev_byte, cur_byte );
    else
      next_price += LZe_price_matched( encoder, prev_byte, cur_byte, match_byte );
    if( !Mf_move_pos( encoder->matchfinder ) ) return 0;

    /* try last updates to next trial */
    next_trial = &encoder->trials[cur+1];

    Tr_update( next_trial, next_price, -1, cur );

    match_price = cur_trial->price + price1( encoder->bm_match[cur_state][pos_state] );
    rep_match_price = match_price + price1( encoder->bm_rep[cur_state] );

    if( match_byte == cur_byte && next_trial->dis != 0 )
      {
      const int price = rep_match_price +
                        LZe_price_rep_len1( encoder, cur_state, pos_state );
      if( price <= next_trial->price )
        {
        next_trial->price = price;
        next_trial->dis = 0;
        next_trial->prev_index = cur;
        next_trial->prev_index2 = single_step_trial;
        }
      }

    available_bytes = min( Mf_available_bytes( encoder->matchfinder ) + 1,
                           max_num_trials - 1 - cur );
    if( available_bytes < min_match_len ) continue;

    len_limit = min( encoder->matchfinder->match_len_limit, available_bytes );

    /* try literal + rep0 */
    if( match_byte != cur_byte && next_trial->prev_index != cur )
      {
      const uint8_t * const data = Mf_ptr_to_current_pos( encoder->matchfinder ) - 1;
      const int dis = cur_trial->reps[0] + 1;
      const int limit = min( encoder->matchfinder->match_len_limit + 1,
                             available_bytes );
      len = 1;
      while( len < limit && data[len-dis] == data[len] ) ++len;
      if( --len >= min_match_len )
        {
        const int pos_state2 = ( pos_state + 1 ) & pos_state_mask;
        const State state2 = St_set_char( cur_state );
        const int price = next_price +
                  price1( encoder->bm_match[state2][pos_state2] ) +
                  price1( encoder->bm_rep[state2] ) +
                  LZe_price_rep0_len( encoder, len, state2, pos_state2 );
        while( num_trials < cur + 1 + len )
          encoder->trials[++num_trials].price = infinite_price;
        Tr_update2( &encoder->trials[cur+1+len], price, 0, cur + 1 );
        }
      }

    /* try rep distances */
    for( rep = 0; rep < num_rep_distances; ++rep )
      {
      const uint8_t * const data = Mf_ptr_to_current_pos( encoder->matchfinder ) - 1;
      int price;
      const int dis = cur_trial->reps[rep] + 1;

      if( data[-dis] != data[0] || data[1-dis] != data[1] ) continue;
      for( len = min_match_len; len < len_limit; ++len )
        if( data[len-dis] != data[len] ) break;
      while( num_trials < cur + len )
        encoder->trials[++num_trials].price = infinite_price;
      price = rep_match_price +
              LZe_price_rep( encoder, rep, cur_state, pos_state );
      for( i = min_match_len; i <= len; ++i )
        Tr_update( &encoder->trials[cur+i], price +
                   Lee_price( &encoder->rep_len_encoder, i, pos_state ),
                   rep, cur );

      if( rep == 0 ) start_len = len + 1;	/* discard shorter matches */

      /* try rep + literal + rep0 */
      {
      int len2 = len + 1, pos_state2;
      const int limit = min( encoder->matchfinder->match_len_limit + len2,
                             available_bytes );
      State state2;
      while( len2 < limit && data[len2-dis] == data[len2] ) ++len2;
      len2 -= len + 1;
      if( len2 < min_match_len ) continue;

      pos_state2 = ( pos_state + len ) & pos_state_mask;
      state2 = St_set_rep( cur_state );
      price += Lee_price( &encoder->rep_len_encoder, len, pos_state ) +
               price0( encoder->bm_match[state2][pos_state2] ) +
               LZe_price_matched( encoder, data[len-1], data[len], data[len-dis] );
      pos_state2 = ( pos_state2 + 1 ) & pos_state_mask;
      state2 = St_set_char( state2 );
      price += price1( encoder->bm_match[state2][pos_state2] ) +
               price1( encoder->bm_rep[state2] ) +
               LZe_price_rep0_len( encoder, len2, state2, pos_state2 );
      while( num_trials < cur + len + 1 + len2 )
        encoder->trials[++num_trials].price = infinite_price;
      Tr_update3( &encoder->trials[cur+len+1+len2], price, 0, cur + len + 1,
                  rep, cur );
      }
      }

    /* try matches */
    if( newlen >= start_len && newlen <= len_limit )
      {
      int dis;
      const int normal_match_price = match_price +
                                     price0( encoder->bm_rep[cur_state] );

      while( num_trials < cur + newlen )
        encoder->trials[++num_trials].price = infinite_price;

      i = 0;
      while( start_len > encoder->pairs[i].len ) ++i;
      dis = encoder->pairs[i].dis;
      for( len = start_len; ; ++len )
        {
        int price = normal_match_price +
                    LZe_price_pair( encoder, dis, len, pos_state );

        Tr_update( &encoder->trials[cur+len], price, dis + num_rep_distances, cur );

        /* try match + literal + rep0 */
        if( len == encoder->pairs[i].len )
          {
          const uint8_t * const data = Mf_ptr_to_current_pos( encoder->matchfinder ) - 1;
          const int dis2 = dis + 1;
          int len2 = len + 1;
          const int limit = min( encoder->matchfinder->match_len_limit + len2,
                                 available_bytes );
          while( len2 < limit && data[len2-dis2] == data[len2] ) ++len2;
          len2 -= len + 1;
          if( len2 >= min_match_len )
            {
            int pos_state2 = ( pos_state + len ) & pos_state_mask;
            State state2 = St_set_match( cur_state );
            price += price0( encoder->bm_match[state2][pos_state2] ) +
                     LZe_price_matched( encoder, data[len-1], data[len], data[len-dis2] );
            pos_state2 = ( pos_state2 + 1 ) & pos_state_mask;
            state2 = St_set_char( state2 );
            price += price1( encoder->bm_match[state2][pos_state2] ) +
                     price1( encoder->bm_rep[state2] ) +
                     LZe_price_rep0_len( encoder, len2, state2, pos_state2 );

            while( num_trials < cur + len + 1 + len2 )
              encoder->trials[++num_trials].price = infinite_price;
            Tr_update3( &encoder->trials[cur+len+1+len2], price, 0,
                        cur + len + 1, dis + num_rep_distances, cur );
            }
          if( ++i >= num_pairs ) break;
          dis = encoder->pairs[i].dis;
          }
        }
      }
    }
  }


static bool LZe_encode_member( struct LZ_encoder * const encoder )
  {
  const int fill_count =
    ( encoder->matchfinder->match_len_limit > 12 ) ? 128 : 512;
  int ahead, i;
  State * const state = &encoder->state;

  if( encoder->member_finished ) return true;
  if( Re_member_position( &encoder->renc ) >= encoder->member_size_limit )
    {
    if( LZe_full_flush( encoder, *state ) ) encoder->member_finished = true;
    return true;
    }

  if( Mf_data_position( encoder->matchfinder ) == 0 &&
      !Mf_finished( encoder->matchfinder ) )	/* encode first byte */
    {
    const uint8_t prev_byte = 0;
    uint8_t cur_byte;
    if( Mf_available_bytes( encoder->matchfinder ) < max_match_len &&
        !Mf_flushing_or_end( encoder->matchfinder ) )
      return true;
    cur_byte = Mf_peek( encoder->matchfinder, 0 );
    Re_encode_bit( &encoder->renc, &encoder->bm_match[*state][0], 0 );
    LZe_encode_literal( encoder, prev_byte, cur_byte );
    CRC32_update_byte( &encoder->crc, cur_byte );
    Mf_get_match_pairs( encoder->matchfinder, 0 );
    if( !Mf_move_pos( encoder->matchfinder ) ) return false;
    }

  while( !Mf_finished( encoder->matchfinder ) )
    {
    if( !Mf_enough_available_bytes( encoder->matchfinder ) ||
        !Re_enough_free_bytes( &encoder->renc ) ) return true;
    if( encoder->pending_num_pairs == 0 )
      {
      if( encoder->fill_counter <= 0 )
        { LZe_fill_distance_prices( encoder ); encoder->fill_counter = fill_count; }
      if( encoder->align_price_count <= 0 )
        LZe_fill_align_prices( encoder );
      }

    ahead = LZe_sequence_optimizer( encoder, encoder->rep_distances, *state );
    if( ahead <= 0 ) return false;		/* can't happen */

    for( i = 0; ; )
      {
      const int pos_state =
        ( Mf_data_position( encoder->matchfinder ) - ahead ) & pos_state_mask;
      const int dis = encoder->trials[i].dis;
      const int len = encoder->trials[i].price;

      bool bit = ( dis < 0 && len == 1 );
      Re_encode_bit( &encoder->renc,
                     &encoder->bm_match[*state][pos_state], !bit );
      if( bit )					/* literal byte */
        {
        const uint8_t prev_byte = Mf_peek( encoder->matchfinder, -ahead-1 );
        const uint8_t cur_byte = Mf_peek( encoder->matchfinder, -ahead );
        CRC32_update_byte( &encoder->crc, cur_byte );
        if( St_is_char( *state ) )
          LZe_encode_literal( encoder, prev_byte, cur_byte );
        else
          {
          const uint8_t match_byte =
            Mf_peek( encoder->matchfinder, -ahead-encoder->rep_distances[0]-1 );
          LZe_encode_matched( encoder, prev_byte, cur_byte, match_byte );
          }
        *state = St_set_char( *state );
        }
      else					/* match or repeated match */
        {
        CRC32_update_buf( &encoder->crc, Mf_ptr_to_current_pos( encoder->matchfinder ) - ahead, len );
        LZe_mtf_reps( dis, encoder->rep_distances );
        bit = ( dis < num_rep_distances );
        Re_encode_bit( &encoder->renc, &encoder->bm_rep[*state], bit );
        if( bit )
          {
          bit = ( dis == 0 );
          Re_encode_bit( &encoder->renc, &encoder->bm_rep0[*state], !bit );
          if( bit )
            Re_encode_bit( &encoder->renc, &encoder->bm_len[*state][pos_state], len > 1 );
          else
            {
            Re_encode_bit( &encoder->renc, &encoder->bm_rep1[*state], dis > 1 );
            if( dis > 1 )
              Re_encode_bit( &encoder->renc, &encoder->bm_rep2[*state], dis > 2 );
            }
          if( len == 1 ) *state = St_set_short_rep( *state );
          else
            {
            Lee_encode( &encoder->rep_len_encoder, &encoder->renc, len, pos_state );
            *state = St_set_rep( *state );
            }
          }
        else
          {
          LZe_encode_pair( encoder, dis - num_rep_distances, len, pos_state );
          --encoder->fill_counter;
          *state = St_set_match( *state );
          }
        }
      ahead -= len; i += len;
      if( Re_member_position( &encoder->renc ) >= encoder->member_size_limit )
        {
        if( !Mf_dec_pos( encoder->matchfinder, ahead ) ) return false;
        if( LZe_full_flush( encoder, *state ) ) encoder->member_finished = true;
        return true;
        }
      if( ahead <= 0 ) break;
      }
    }
  if( LZe_full_flush( encoder, *state ) ) encoder->member_finished = true;
  return true;
  }