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
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
// Copyright 2023 The rust-ggstd authors. All rights reserved.
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.

//! Package flate implements the DEFLATE compressed data format, described in
//! RFC 1951.  The gzip and zlib packages implement access to DEFLATE-based file
//! formats.

use super::deflate::MAX_MATCH_OFFSET;
use super::dict_decoder::DictDecoder;
use super::huffman_bit_writer::END_BLOCK_MARKER;
use crate::compat;
use crate::errors;
use crate::io as ggio;
use crate::math::bits;
use std::sync::OnceLock;

// ggrust TODO: try to get rid of self.err when all tests are passing

const MAX_CODE_LEN: usize = 16; // max length of Huffman code

// The next three numbers come from the RFC section 3.2.7, with the
// additional proviso in section 3.2.5 which implies that distance codes
// 30 and 31 should never occur in compressed data.
pub(super) const MAX_NUM_LIT: usize = 286;
const MAX_NUM_DIST: usize = 30;
const NUM_CODES: usize = 19; // number of codes in Huffman meta-code

// var fixedOnce sync.Once
// var fixedHuffmanDecoder HuffmanDecoder
/// Initialize the fixedHuffmanDecoder only once upon first use.
pub(super) fn get_fixed_huffman_decoder() -> &'static HuffmanDecoder {
    static ENCODER: OnceLock<HuffmanDecoder> = OnceLock::new();
    ENCODER.get_or_init(fixed_huffman_decoder_init)
}

// // Resetter resets a ReadCloser returned by new_reader or new_reader_dict
// // to switch to a new underlying Reader. This permits reusing a ReadCloser
// // instead of allocating a new one.
// type Resetter interface {
// 	// reset discards any buffered data and resets the Resetter as if it was
// 	// newly initialized with the given reader.
// 	reset(r ggio::Reader, dict : &[u8]) error
// }

// The data structure for decoding Huffman tables is based on that of
// zlib. There is a lookup table of a fixed bit width (huffmanChunkBits),
// For codes smaller than the table width, there are multiple entries
// (each combination of trailing bits has the same value). For codes
// larger than the table width, the table contains a link to an overflow
// table. The width of each entry in the link table is the maximum code
// size minus the chunk width.
//
// Note that you can do a lookup in the table even without all bits
// filled. Since the extra bits are zero, and the DEFLATE Huffman codes
// have the property that shorter codes come before longer ones, the
// bit length estimate in the result is a lower bound on the actual
// number of bits.
//
// See the following:
//	https://github.com/madler/zlib/raw/master/doc/algorithm.txt

// chunk & 15 is number of bits
// chunk >> 4 is value, including table link

const HUFFMAN_CHUNK_BITS: u32 = 9;
const HUFFMAN_NUM_CHUNKS: u32 = 1 << HUFFMAN_CHUNK_BITS;
const HUFFMAN_COUNT_MASK: u32 = 15;
const HUFFMAN_VALUE_SHIFT: u32 = 4;

pub(super) struct HuffmanDecoder {
    min: u32,             // the minimum code length
    chunks: Vec<u32>,     // [huffmanNumChunks]u32 // chunks as described above
    links: Vec<Vec<u32>>, // [][]u32               // overflow links
    link_mask: u32,       // mask the width of the link table
}

impl HuffmanDecoder {
    pub(super) fn new() -> Self {
        Self {
            min: 0,
            chunks: vec![0; HUFFMAN_NUM_CHUNKS as usize],
            links: Vec::new(),
            link_mask: 0,
        }
    }

    /// Initialize Huffman decoding tables from array of code lengths.
    /// Following this function, h is guaranteed to be initialized into a complete
    /// tree (i.e., neither over-subscribed nor under-subscribed). The exception is a
    /// degenerate case where the tree has only a single symbol with length 1. Empty
    /// trees are permitted.
    pub(super) fn init(&mut self, lengths: &[u32]) -> bool {
        // Sanity enables additional runtime tests during Huffman
        // table construction. It's intended to be used during
        // development to supplement the currently ad-hoc unit tests.
        const SANITY: bool = false;
        // const SANITY: bool = true;

        if self.min != 0 {
            self.min = 0;
            self.chunks = vec![0; HUFFMAN_NUM_CHUNKS as usize];
            self.links = Vec::new();
            self.link_mask = 0;
        }

        // Count number of codes of each length,
        // compute min and max length.
        let mut count = [0_u32; MAX_CODE_LEN];
        let mut min: u32 = 0;
        let mut max: u32 = 0;
        for n in lengths {
            let n = *n;
            if n == 0 {
                continue;
            }
            if min == 0 || n < min {
                min = n;
            }
            if n > max {
                max = n;
            }
            count[n as usize] += 1;
        }

        // Empty tree. The decompressor.huff_sym function will fail later if the tree
        // is used. Technically, an empty tree is only valid for the HDIST tree and
        // not the HCLEN and HLIT tree. However, a stream with an empty HCLEN tree
        // is guaranteed to fail since it will attempt to use the tree to decode the
        // codes for the HLIT and HDIST trees. Similarly, an empty HLIT tree is
        // guaranteed to fail later since the compressed data section must be
        // composed of at least one symbol (the end-of-block marker).
        if max == 0 {
            return true;
        }

        let mut code = 0_u32;
        let mut nextcode = [0_u32; MAX_CODE_LEN];
        for i in min..=max {
            code <<= 1;
            nextcode[i as usize] = code;
            code += count[i as usize];
        }

        // Check that the coding is complete (i.e., that we've
        // assigned all 2-to-the-max possible bit sequences).
        // Exception: To be compatible with zlib, we also need to
        // accept degenerate single-code codings. See also
        // TestDegenerateHuffmanCoding.
        if code != 1 << (max as usize) && !(code == 1 && max == 1) {
            return false;
        }

        self.min = min;
        if max > HUFFMAN_CHUNK_BITS {
            let num_links = 1_u32 << (max - HUFFMAN_CHUNK_BITS);
            self.link_mask = num_links - 1;

            // create link tables
            let link = nextcode[HUFFMAN_CHUNK_BITS as usize + 1] >> 1;
            self.links = vec![Vec::new(); (HUFFMAN_NUM_CHUNKS - link) as usize];
            for j in link..HUFFMAN_NUM_CHUNKS {
                let mut reverse = bits::reverse16(j as u16);
                reverse >>= 16 - HUFFMAN_CHUNK_BITS;
                let off = j - link;
                if SANITY && self.chunks[reverse as usize] != 0 {
                    panic!("impossible: overwriting existing chunk");
                }
                self.chunks[reverse as usize] =
                    (off << HUFFMAN_VALUE_SHIFT) | (HUFFMAN_CHUNK_BITS + 1);
                self.links[off as usize] = vec![0; num_links as usize];
            }
        }

        for (i, n) in lengths.iter().enumerate() {
            let n = *n;
            if n == 0 {
                continue;
            }
            let code = nextcode[n as usize];
            nextcode[n as usize] += 1;
            let chunk = ((i as u32) << HUFFMAN_VALUE_SHIFT) | n;
            let mut reverse = bits::reverse16(code as u16) as u32;
            reverse >>= 16 - n;
            if n <= HUFFMAN_CHUNK_BITS {
                let mut off = reverse;
                while off < self.chunks.len() as u32 {
                    // 			for off := reverse; off < len(self.chunks); off += 1 << usize(n) {
                    // We should never need to overwrite
                    // an existing chunk. Also, 0 is
                    // never a valid chunk, because the
                    // lower 4 "count" bits should be
                    // between 1 and 15.
                    if SANITY && self.chunks[off as usize] != 0 {
                        panic!("impossible: overwriting existing chunk")
                    }
                    self.chunks[off as usize] = chunk;
                    off += 1 << n;
                }
            } else {
                let j = reverse & (HUFFMAN_NUM_CHUNKS - 1);
                if SANITY
                    && (self.chunks[j as usize] & HUFFMAN_COUNT_MASK) != HUFFMAN_CHUNK_BITS + 1
                {
                    // Longer codes should have been
                    // associated with a link table above.
                    panic!("impossible: not an indirect chunk");
                }
                let value = self.chunks[j as usize] >> HUFFMAN_VALUE_SHIFT;
                let linktab = &mut self.links[value as usize];
                reverse >>= HUFFMAN_CHUNK_BITS;
                let mut off = reverse;
                while off < linktab.len() as u32 {
                    if SANITY && linktab[off as usize] != 0 {
                        panic!("impossible: overwriting existing chunk");
                    }
                    linktab[off as usize] = chunk;
                    off += 1 << (n - HUFFMAN_CHUNK_BITS);
                }
            }
        }

        if SANITY {
            // Above we've sanity checked that we never overwrote
            // an existing entry. Here we additionally check that
            // we filled the tables completely.
            for (i, chunk) in self.chunks.iter().enumerate() {
                let chunk = *chunk;
                if chunk == 0 {
                    // As an exception, in the degenerate
                    // single-code case, we allow odd
                    // chunks to be missing.
                    if code == 1 && i % 2 == 1 {
                        continue;
                    }
                    panic!("impossible: missing chunk")
                }
            }
            for linktab in self.links.iter() {
                for chunk in linktab {
                    if *chunk == 0 {
                        panic!("impossible: missing chunk");
                    }
                }
            }
        }

        true
    }
}

#[derive(PartialEq)]
enum HDDecoder {
    None,
    H2,
}

#[derive(PartialEq)]
enum HLDecoder {
    Fixed,
    H1,
}

/// Reader uncompress data provided by another reader (input source).
/// Reader returns Ok(0) after the final block in the DEFLATE stream has
/// been encountered. Any trailing data after the final block is ignored.
///
/// The input source must be buffered because the implementation uses
/// many 1-byte reads.
pub struct Reader<Input: std::io::BufRead> {
    /// Input source.
    r: Input,
    roffset: u64,
    // Input bits, in top of b.
    b: u32,
    nb: usize,
    // Huffman decoders for literal/length, distance.
    pub(super) h1: HuffmanDecoder,
    h2: HuffmanDecoder,
    // Length arrays used to define Huffman codes.
    bits: Vec<u32>, // [isize; MAX_NUM_LIT + MAX_NUM_DIST],
    codebits: [u32; NUM_CODES],

    // Output history, buffer.
    pub(super) dict: DictDecoder,

    // Temporary buffer (avoids repeated allocation).
    buf: [u8; 4],
    // Next step in the decompression,
    // and decompression state.
    step: StepFunc,
    step_state: StepState,
    final_: bool,
    end_of_stream: bool,
    err: Option<std::io::Error>,
    hl: HLDecoder,
    hd: HDDecoder,
    copy_len: usize,
    copy_dist: usize,
}

enum StepFunc {
    NextBlock,
    HuffmanBlock,
    CopyData,
}

#[derive(PartialEq, Copy, Clone)]
enum StepState {
    StateInit,
    StateDict,
}

pub(super) enum DecoderToUse {
    H1,
    HL,
    H2,
}

// RFC 1951 section 3.2.7.
// Compression with dynamic Huffman codes

const CODE_ORDER: [usize; 19] = [
    16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15,
];

impl<Input: std::io::BufRead> std::io::Read for Reader<Input> {
    fn read(&mut self, buf: &mut [u8]) -> std::io::Result<usize> {
        errors::iores_to_result(crate::io::Reader::read(self, buf))
    }
}

impl<'a, Input: std::io::BufRead> Reader<Input> {
    /// new returns a new Reader that can be used
    /// to read the uncompressed version of r.
    /// The reader returns std::io::Error::EOF after the final block in the DEFLATE stream has
    /// been encountered. Any trailing data after the final block is ignored.
    ///
    // ggrust: not implemented
    // The ReadCloser returned by new_reader also implements Resetter.
    pub fn new(r: Input) -> Reader<Input> {
        Self::new_dict(r, &[])
    }

    /// new_dict is like new but initializes the reader
    /// with a preset dictionary. The returned Reader behaves as if
    /// the uncompressed data stream started with the given dictionary,
    /// which has already been read. new_dict is typically used
    /// to read data compressed by Writer::new_dict.
    ///
    // ggrust: not implemented
    // The ReadCloser returned by new_reader also implements Resetter.
    pub fn new_dict(r: Input, dict: &'a [u8]) -> Reader<Input> {
        Reader {
            r,
            roffset: 0,
            b: 0,
            nb: 0,
            h1: HuffmanDecoder::new(),
            h2: HuffmanDecoder::new(),
            bits: vec![0; MAX_NUM_LIT + MAX_NUM_DIST],
            codebits: [0; NUM_CODES],
            dict: DictDecoder::new(MAX_MATCH_OFFSET, dict),
            buf: [0; 4],
            step: StepFunc::NextBlock,
            step_state: StepState::StateInit,
            final_: false,
            end_of_stream: false,
            err: None,
            hl: HLDecoder::Fixed,
            hd: HDDecoder::None,
            copy_len: 0,
            copy_dist: 0,
        }
    }

    fn next_block(&mut self) {
        while self.nb < 1 + 2 {
            let res = self.more_bits();
            if res.is_err() {
                self.err = Some(res.err().unwrap());
                return;
            }
        }
        self.final_ = self.b & 1 == 1;
        self.b >>= 1;
        let typ = self.b & 3;
        self.b >>= 2;
        self.nb -= 1 + 2;
        match typ {
            0 => self.data_block(),
            1 => {
                // compressed, fixed Huffman tables
                self.hl = HLDecoder::Fixed;
                self.hd = HDDecoder::None;
                self.huffman_block();
            }
            2 => {
                // compressed, dynamic Huffman tables
                let res = self.read_huffman();
                if res.is_ok() {
                    self.hl = HLDecoder::H1;
                    self.hd = HDDecoder::H2;
                    self.huffman_block();
                }
            }
            _ => {
                // 3 is reserved.
                self.err = Some(new_corrupted_input_error(self.roffset));
            }
        }
    }

    pub fn close(&mut self) -> std::io::Result<()> {
        // if self.err.is_some() {
        //     let err = self.err.as_ref().unwrap();
        //     if err.is_eof() {
        //         return Ok(());
        //     }
        //     return Err(err.lossy_copy());
        // }
        Ok(())
    }

    fn read_huffman(&mut self) -> std::io::Result<()> {
        // HLIT[5], HDIST[5], HCLEN[4].
        while self.nb < 5 + 5 + 4 {
            self.more_bits()?
        }
        let nlit = (self.b & 0x1F) as usize + 257;
        if nlit > MAX_NUM_LIT {
            return Err(new_corrupted_input_error(self.roffset));
        }
        self.b >>= 5;
        let ndist = (self.b & 0x1F) as usize + 1;
        if ndist > MAX_NUM_DIST {
            return Err(new_corrupted_input_error(self.roffset));
        }
        self.b >>= 5;
        let nclen = (self.b & 0xF) as usize + 4;
        // NUM_CODES is 19, so nclen is always valid.
        self.b >>= 4;
        self.nb -= 5 + 5 + 4;

        // (HCLEN+4)*3 bits: code lengths in the magic CODE_ORDER order.
        #[allow(clippy::needless_range_loop)]
        for i in 0..nclen {
            while self.nb < 3 {
                self.more_bits()?
            }
            self.codebits[CODE_ORDER[i]] = self.b & 0x7;
            self.b >>= 3;
            self.nb -= 3;
        }
        #[allow(clippy::needless_range_loop)]
        for i in nclen..CODE_ORDER.len() {
            self.codebits[CODE_ORDER[i]] = 0;
        }
        if !self.h1.init(&self.codebits[0..]) {
            return Err(new_corrupted_input_error(self.roffset));
        }

        // HLIT + 257 code lengths, HDIST + 1 code lengths,
        // using the code length Huffman code.
        let mut i = 0;
        let n = nlit + ndist;
        while i < n {
            let x = self.huff_sym(DecoderToUse::H1)?;
            if x < 16 {
                // Actual length.
                self.bits[i] = x as u32;
                i += 1;
                continue;
            }
            // Repeat previous length or zero.
            let mut rep: usize;
            let nb: usize;
            let b: u32;
            match x {
                16 => {
                    rep = 3;
                    nb = 2;
                    if i == 0 {
                        return Err(new_corrupted_input_error(self.roffset));
                    }
                    b = self.bits[i - 1];
                }
                17 => {
                    rep = 3;
                    nb = 3;
                    b = 0;
                }
                18 => {
                    rep = 11;
                    nb = 7;
                    b = 0;
                }
                _ => {
                    return Err(std::io::Error::new(
                        std::io::ErrorKind::InvalidData,
                        "unexpected length code",
                    ))
                }
            };
            while self.nb < nb {
                self.more_bits()?
            }
            rep += (self.b & ((1 << nb) - 1) as u32) as usize;
            self.b >>= nb;
            self.nb -= nb;
            if i + rep > n {
                return Err(new_corrupted_input_error(self.roffset));
            }
            for _j in 0..rep {
                self.bits[i] = b;
                i += 1;
            }
        }

        if !self.h1.init(&self.bits[0..nlit]) || !self.h2.init(&self.bits[nlit..nlit + ndist]) {
            return Err(new_corrupted_input_error(self.roffset));
        }

        // As an optimization, we can initialize the min bits to read at a time
        // for the HLIT tree to the length of the EOB marker since we know that
        // every block must terminate with one. This preserves the property that
        // we never read any extra bytes after the end of the DEFLATE stream.
        if self.h1.min < self.bits[END_BLOCK_MARKER] {
            self.h1.min = self.bits[END_BLOCK_MARKER];
        }
        Ok(())
    }

    // Decode a single Huffman block from self.
    // hl and hd are the Huffman states for the lit/length values
    // and the distance values, respectively. If hd == nil, using the
    // fixed distance encoding associated with fixed Huffman blocks.
    fn huffman_block(&mut self) {
        enum StateMachine {
            ReadLiteral,
            CopyHistory,
        }
        let mut next_step = match self.step_state {
            StepState::StateInit => StateMachine::ReadLiteral,
            StepState::StateDict => StateMachine::CopyHistory,
        };
        loop {
            match next_step {
                StateMachine::ReadLiteral => {
                    // Read literal and/or (length, distance) according to RFC section 3.2.3.
                    let v = match self.huff_sym(DecoderToUse::HL) {
                        Ok(v) => v,
                        Err(err) => {
                            self.err = Some(err);
                            return;
                        }
                    };
                    let n: usize; // number of bits extra
                    let length: usize;
                    match v {
                    0..=255 => {
                        // 		case v < 256:
                        self.dict.write_byte(v as u8);
                        if self.dict.avail_write() == 0 {
                            self.dict.stash_flush();
                            self.step = StepFunc::HuffmanBlock;
                            self.step_state = StepState::StateInit;
                            return;
                        }
                        // next_step = StateMachine::ReadLiteral;
                        continue;
                    }
                    256 => {
                        // 		case v == 256:
                        self.finish_block();
                        return;
                    }
                    // otherwise, reference to older data
                    257..=264 => {
                        // 		case v < 265:
                        length = v - (257 - 3);
                        n = 0;
                    }
                    265..=268 => {
                        // 		case v < 269:
                        length = v*2 - (265*2 - 11);
                        n = 1;
                    }
                    269..=272 => {
                        // 		case v < 273:
                        length = v*4 - (269*4 - 19);
                        n = 2;
                    }
                    273..=276 => {
                        // 		case v < 277:
                        length = v*8 - (273*8 - 35);
                        n = 3;
                    }
                    277..=280 => {
                        // 		case v < 281:
                        length = v*16 - (277*16 - 67);
                        n = 4;
                    }
                    281..=284 => {
                        // 		case v < 285:
                        length = v*32 - (281*32 - 131);
                        n = 5;
                    }
                    285..=285 /* MAX_NUM_LIT-1 */ => {
                        // 		case v < MAX_NUM_LIT:
                        length = 258;
                        n = 0;
                    }
                    _ => {
                        self.err = Some(new_corrupted_input_error(self.roffset));
                        return;
                    }
                };
                    let mut length = length;
                    if n > 0 {
                        while self.nb < n {
                            if let Err(err) = self.more_bits() {
                                self.err = Some(err);
                                return;
                            }
                        }
                        length += (self.b & ((1 << n) - 1) as u32) as usize;
                        self.b >>= n;
                        self.nb -= n;
                    }

                    let mut dist: usize;
                    match self.hd {
                        HDDecoder::None => {
                            while self.nb < 5 {
                                if let Err(err) = self.more_bits() {
                                    self.err = Some(err);
                                    return;
                                }
                            }
                            dist = bits::reverse8(((self.b & 0x1F) << 3) as u8) as usize;
                            self.b >>= 5;
                            self.nb -= 5;
                        }
                        HDDecoder::H2 => {
                            let res = self.huff_sym(DecoderToUse::H2);
                            dist = match res {
                                Ok(v) => v,
                                Err(err) => {
                                    self.err = Some(err);
                                    return;
                                }
                            };
                        }
                    }

                    if (0..4).contains(&dist) {
                        // 		case dist < 4:
                        dist += 1;
                    } else if (4..MAX_NUM_DIST).contains(&dist) {
                        // 		case dist < MAX_NUM_DIST:
                        let nb = (dist - 2) >> 1;
                        // have 1 bit in bottom of dist, need nb more.
                        let mut extra = (dist & 1) << nb;
                        while self.nb < nb {
                            if let Err(err) = self.more_bits() {
                                self.err = Some(err);
                                return;
                            }
                        }
                        extra |= (self.b & ((1 << nb) - 1) as u32) as usize;
                        self.b >>= nb;
                        self.nb -= nb;
                        dist = 1_usize.overflowing_shl((nb + 1) as u32).0 + 1 + extra;
                    } else {
                        // 		default:
                        self.err = Some(new_corrupted_input_error(self.roffset));
                        return;
                    }

                    // No check on length; encoding can be prescient.
                    if dist > self.dict.hist_size() {
                        self.err = Some(new_corrupted_input_error(self.roffset));
                        return;
                    }

                    self.copy_len = length;
                    self.copy_dist = dist;
                    next_step = StateMachine::CopyHistory;
                }
                StateMachine::CopyHistory => {
                    // Perform a backwards copy according to RFC section 3.2.3.
                    let mut cnt = self.dict.try_write_copy(self.copy_dist, self.copy_len);
                    if cnt == 0 {
                        cnt = self.dict.write_copy(self.copy_dist, self.copy_len);
                    }
                    self.copy_len -= cnt;

                    if self.dict.avail_write() == 0 || self.copy_len > 0 {
                        self.dict.stash_flush();
                        self.step = StepFunc::HuffmanBlock; // We need to continue this work
                        self.step_state = StepState::StateDict;
                        return;
                    }
                    next_step = StateMachine::ReadLiteral;
                }
            }
        }
    }

    /// Copy a single uncompressed data block from input to output.
    fn data_block(&mut self) {
        // Uncompressed.
        // Discard current half-u8.
        self.nb = 0;
        self.b = 0;

        // Length then ones-complement of length.
        let (n, err) = ggio::read_full(&mut self.r, &mut self.buf[0..4]);
        self.roffset += n as u64;
        if let Some(err) = err {
            self.err = Some(err);
            return;
        }
        let n = (self.buf[0]) as usize | ((self.buf[1] as usize) << 8);
        let nn = (self.buf[2]) as usize | ((self.buf[3] as usize) << 8);
        if (nn as u16) != (!n as u16) {
            self.err = Some(new_corrupted_input_error(self.roffset));
            return;
        }

        if n == 0 {
            self.dict.stash_flush();
            self.finish_block();
            return;
        }

        self.copy_len = n;
        self.copy_data();
    }

    /// copy_data copies self.copy_len bytes from the underlying reader into self.hist.
    /// It pauses for reads when self.hist is full.
    fn copy_data(&mut self) {
        let buf = self.dict.write_slice(self.copy_len);
        let (n, err) = ggio::read_full(&mut self.r, buf);
        self.roffset += n as u64;
        self.copy_len -= n;
        self.dict.write_mark(n);
        if let Some(err) = err {
            self.err = Some(err);
            return;
        }

        if self.dict.avail_write() == 0 || self.copy_len > 0 {
            self.dict.stash_flush();
            self.step = StepFunc::CopyData;
            return;
        }
        self.finish_block();
    }

    fn finish_block(&mut self) {
        if self.final_ {
            if self.dict.avail_read() > 0 {
                self.dict.stash_flush();
            }
            self.end_of_stream = true;
        }
        self.step = StepFunc::NextBlock;
    }

    fn more_bits(&mut self) -> std::io::Result<()> {
        let c = compat::readers::read_byte(&mut self.r)?;
        self.roffset += 1;
        self.b |= (c as u32) << self.nb;
        self.nb += 8;
        Ok(())
    }

    /// Read the next Huffman-encoded symbol from f according to h.
    pub(super) fn huff_sym(&mut self, decoder: DecoderToUse) -> Result<usize, std::io::Error> {
        let h = match decoder {
            DecoderToUse::H1 => &self.h1,
            DecoderToUse::HL => match self.hl {
                HLDecoder::Fixed => get_fixed_huffman_decoder(),
                HLDecoder::H1 => &self.h1,
            },
            DecoderToUse::H2 => &self.h2,
        };
        // Since a HuffmanDecoder can be empty or be composed of a degenerate tree
        // with single element, huff_sym must error on these two edge cases. In both
        // cases, the chunks slice will be 0 for the invalid sequence, leading it
        // satisfy the n == 0 check below.
        let mut n = h.min as usize;
        // Optimization. Compiler isn't smart enough to keep self.b,self.nb in registers,
        // but is smart enough to keep local variables in registers, so use nb and b,
        // inline call to more_bits and reassign b,nb back to f on return.
        let (mut nb, mut b) = (self.nb, self.b);
        loop {
            while nb < n {
                match compat::readers::read_byte(&mut self.r) {
                    Err(err) => {
                        self.b = b;
                        self.nb = nb;
                        return Err(err);
                    }
                    Ok(c) => {
                        self.roffset += 1;
                        b |= (c as u32) << (nb & 31);
                        nb = nb.wrapping_add(8);
                    }
                }
            }
            let mut chunk = h.chunks[(b & (HUFFMAN_NUM_CHUNKS - 1)) as usize];
            n = (chunk & HUFFMAN_COUNT_MASK) as usize;
            if n > HUFFMAN_CHUNK_BITS as usize {
                chunk = h.links[(chunk >> HUFFMAN_VALUE_SHIFT) as usize]
                    [((b >> HUFFMAN_CHUNK_BITS) & h.link_mask) as usize];
                n = (chunk & HUFFMAN_COUNT_MASK) as usize;
            }
            if n <= nb {
                if n == 0 {
                    self.b = b;
                    self.nb = nb;
                    self.err = Some(new_corrupted_input_error(self.roffset));
                    return Err(errors::copy_stdio_error(self.err.as_ref().unwrap()));
                }
                self.b = b >> (n & 31);
                self.nb = nb - n;
                return Ok((chunk >> HUFFMAN_VALUE_SHIFT) as usize);
            }
        }
    }

    // Reset discards any buffered data and resets the instance if it was
    // newly initialized with the given reader.
    pub fn reset(&mut self, r: Input, dict: &[u8]) {
        self.r = r;
        self.reset_state(dict);
    }

    // reset_state discards any buffered data and resets the instance if it was
    // newly initialized with the reader it already holds.
    pub fn reset_state(&mut self, dict: &[u8]) {
        // keep following fields
        // self.bits = vec![0; MAX_NUM_LIT + MAX_NUM_DIST];
        // self.codebits = [0; NUM_CODES];

        // reset everything else
        self.dict = DictDecoder::new(MAX_MATCH_OFFSET, dict);

        self.roffset = 0;
        self.b = 0;
        self.nb = 0;
        self.h1 = HuffmanDecoder::new();
        self.h2 = HuffmanDecoder::new();
        self.buf = [0; 4];
        self.step = StepFunc::NextBlock;
        self.step_state = StepState::StateInit;
        self.final_ = false;
        self.end_of_stream = false;
        self.err = None;
        self.hl = HLDecoder::Fixed;
        self.hd = HDDecoder::None;
        self.copy_len = 0;
        self.copy_dist = 0;
    }

    /// input_reader returns mutable reference to the underlying reader.
    pub fn input_reader(&mut self) -> &mut Input {
        &mut self.r
    }
}

impl<Input: std::io::BufRead> crate::io::Reader for Reader<Input> {
    fn read(&mut self, p: &mut [u8]) -> ggio::IoRes {
        // first returning available data, then an error if it exists
        loop {
            if self.dict.stash_len() > 0 {
                let n = self.dict.stash_read(p);
                if n > 0 {
                    return (n, None);
                }
            }
            if self.err.is_some() {
                return (0, errors::copy_stdio_option_error(&self.err));
            }
            if self.end_of_stream {
                return ggio::EOF;
            }
            match self.step {
                StepFunc::NextBlock => self.next_block(),
                StepFunc::HuffmanBlock => self.huffman_block(),
                StepFunc::CopyData => self.copy_data(),
            }
            if self.err.is_some() && self.dict.stash_len() == 0 {
                self.dict.stash_flush(); // Flush what's left in case of error
            }
        }
    }
}

fn fixed_huffman_decoder_init() -> HuffmanDecoder {
    let mut h = HuffmanDecoder::new();
    // These come from the RFC section 3.2.6.
    let mut bits = [0; 288];
    #[allow(clippy::needless_range_loop)]
    for i in 0..144 {
        bits[i] = 8;
    }
    #[allow(clippy::needless_range_loop)]
    for i in 144..256 {
        bits[i] = 9;
    }
    #[allow(clippy::needless_range_loop)]
    for i in 256..280 {
        bits[i] = 7;
    }
    #[allow(clippy::needless_range_loop)]
    for i in 280..288 {
        bits[i] = 8;
    }
    h.init(&bits);
    h
}

fn new_corrupted_input_error(offset: u64) -> std::io::Error {
    let msg = format!("flate: corrupt input before offset {}", offset);
    std::io::Error::new(std::io::ErrorKind::InvalidData, msg)
}