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
use std::cmp;
use std::io;
use std::io::prelude::*;

use super::{corrupt, GzBuilder, GzHeader, GzHeaderParser};
use crate::crc::{Crc, CrcWriter};
use crate::zio;
use crate::{Compress, Compression, Decompress, Status};

/// A gzip streaming encoder
///
/// This structure exposes a [`Write`] interface that will emit compressed data
/// to the underlying writer `W`.
///
/// [`Write`]: https://doc.rust-lang.org/std/io/trait.Write.html
///
/// # Examples
///
/// ```
/// use std::io::prelude::*;
/// use flate2::Compression;
/// use flate2::write::GzEncoder;
///
/// // Vec<u8> implements Write to print the compressed bytes of sample string
/// # fn main() {
///
/// let mut e = GzEncoder::new(Vec::new(), Compression::default());
/// e.write_all(b"Hello World").unwrap();
/// println!("{:?}", e.finish().unwrap());
/// # }
/// ```
#[derive(Debug)]
pub struct GzEncoder<W: Write> {
    inner: zio::Writer<W, Compress>,
    crc: Crc,
    crc_bytes_written: usize,
    header: Vec<u8>,
}

pub fn gz_encoder<W: Write>(header: Vec<u8>, w: W, lvl: Compression) -> GzEncoder<W> {
    GzEncoder {
        inner: zio::Writer::new(w, Compress::new(lvl, false)),
        crc: Crc::new(),
        header,
        crc_bytes_written: 0,
    }
}

impl<W: Write> GzEncoder<W> {
    /// Creates a new encoder which will use the given compression level.
    ///
    /// The encoder is not configured specially for the emitted header. For
    /// header configuration, see the `GzBuilder` type.
    ///
    /// The data written to the returned encoder will be compressed and then
    /// written to the stream `w`.
    pub fn new(w: W, level: Compression) -> GzEncoder<W> {
        GzBuilder::new().write(w, level)
    }

    /// Acquires a reference to the underlying writer.
    pub fn get_ref(&self) -> &W {
        self.inner.get_ref()
    }

    /// Acquires a mutable reference to the underlying writer.
    ///
    /// Note that mutation of the writer may result in surprising results if
    /// this encoder is continued to be used.
    pub fn get_mut(&mut self) -> &mut W {
        self.inner.get_mut()
    }

    /// Attempt to finish this output stream, writing out final chunks of data.
    ///
    /// Note that this function can only be used once data has finished being
    /// written to the output stream. After this function is called then further
    /// calls to `write` may result in a panic.
    ///
    /// # Panics
    ///
    /// Attempts to write data to this stream may result in a panic after this
    /// function is called.
    ///
    /// # Errors
    ///
    /// This function will perform I/O to complete this stream, and any I/O
    /// errors which occur will be returned from this function.
    pub fn try_finish(&mut self) -> io::Result<()> {
        self.write_header()?;
        self.inner.finish()?;

        while self.crc_bytes_written < 8 {
            let (sum, amt) = (self.crc.sum(), self.crc.amount());
            let buf = [
                (sum >> 0) as u8,
                (sum >> 8) as u8,
                (sum >> 16) as u8,
                (sum >> 24) as u8,
                (amt >> 0) as u8,
                (amt >> 8) as u8,
                (amt >> 16) as u8,
                (amt >> 24) as u8,
            ];
            let inner = self.inner.get_mut();
            let n = inner.write(&buf[self.crc_bytes_written..])?;
            self.crc_bytes_written += n;
        }
        Ok(())
    }

    /// Finish encoding this stream, returning the underlying writer once the
    /// encoding is done.
    ///
    /// Note that this function may not be suitable to call in a situation where
    /// the underlying stream is an asynchronous I/O stream. To finish a stream
    /// the `try_finish` (or `shutdown`) method should be used instead. To
    /// re-acquire ownership of a stream it is safe to call this method after
    /// `try_finish` or `shutdown` has returned `Ok`.
    ///
    /// # Errors
    ///
    /// This function will perform I/O to complete this stream, and any I/O
    /// errors which occur will be returned from this function.
    pub fn finish(mut self) -> io::Result<W> {
        self.try_finish()?;
        Ok(self.inner.take_inner())
    }

    fn write_header(&mut self) -> io::Result<()> {
        while !self.header.is_empty() {
            let n = self.inner.get_mut().write(&self.header)?;
            self.header.drain(..n);
        }
        Ok(())
    }
}

impl<W: Write> Write for GzEncoder<W> {
    fn write(&mut self, buf: &[u8]) -> io::Result<usize> {
        assert_eq!(self.crc_bytes_written, 0);
        self.write_header()?;
        let n = self.inner.write(buf)?;
        self.crc.update(&buf[..n]);
        Ok(n)
    }

    fn flush(&mut self) -> io::Result<()> {
        assert_eq!(self.crc_bytes_written, 0);
        self.write_header()?;
        self.inner.flush()
    }
}

impl<R: Read + Write> Read for GzEncoder<R> {
    fn read(&mut self, buf: &mut [u8]) -> io::Result<usize> {
        self.get_mut().read(buf)
    }
}

impl<W: Write> Drop for GzEncoder<W> {
    fn drop(&mut self) {
        if self.inner.is_present() {
            let _ = self.try_finish();
        }
    }
}

/// A decoder for a single member of a [gzip file].
///
/// This structure exposes a [`Write`] interface, receiving compressed data and
/// writing uncompressed data to the underlying writer.
///
/// After decoding a single member of the gzip data this writer will return the number of bytes up to
/// to the end of the gzip member and subsequent writes will return Ok(0) allowing the caller to
/// handle any data following the gzip member.
///
/// To handle gzip files that may have multiple members, see [`MultiGzDecoder`]
/// or read more
/// [in the introduction](../index.html#about-multi-member-gzip-files).
///
/// [gzip file]: https://www.rfc-editor.org/rfc/rfc1952#page-5
/// [`Write`]: https://doc.rust-lang.org/std/io/trait.Write.html
///
/// # Examples
///
/// ```
/// use std::io::prelude::*;
/// use std::io;
/// use flate2::Compression;
/// use flate2::write::{GzEncoder, GzDecoder};
///
/// # fn main() {
/// #    let mut e = GzEncoder::new(Vec::new(), Compression::default());
/// #    e.write(b"Hello World").unwrap();
/// #    let bytes = e.finish().unwrap();
/// #    assert_eq!("Hello World", decode_writer(bytes).unwrap());
/// # }
/// // Uncompresses a gzip encoded vector of bytes and returns a string or error
/// // Here Vec<u8> implements Write
/// fn decode_writer(bytes: Vec<u8>) -> io::Result<String> {
///    let mut writer = Vec::new();
///    let mut decoder = GzDecoder::new(writer);
///    decoder.write_all(&bytes[..])?;
///    writer = decoder.finish()?;
///    let return_string = String::from_utf8(writer).expect("String parsing error");
///    Ok(return_string)
/// }
/// ```
#[derive(Debug)]
pub struct GzDecoder<W: Write> {
    inner: zio::Writer<CrcWriter<W>, Decompress>,
    crc_bytes: Vec<u8>,
    header_parser: GzHeaderParser,
}

const CRC_BYTES_LEN: usize = 8;

impl<W: Write> GzDecoder<W> {
    /// Creates a new decoder which will write uncompressed data to the stream.
    ///
    /// When this encoder is dropped or unwrapped the final pieces of data will
    /// be flushed.
    pub fn new(w: W) -> GzDecoder<W> {
        GzDecoder {
            inner: zio::Writer::new(CrcWriter::new(w), Decompress::new(false)),
            crc_bytes: Vec::with_capacity(CRC_BYTES_LEN),
            header_parser: GzHeaderParser::new(),
        }
    }

    /// Returns the header associated with this stream.
    pub fn header(&self) -> Option<&GzHeader> {
        self.header_parser.header()
    }

    /// Acquires a reference to the underlying writer.
    pub fn get_ref(&self) -> &W {
        self.inner.get_ref().get_ref()
    }

    /// Acquires a mutable reference to the underlying writer.
    ///
    /// Note that mutating the output/input state of the stream may corrupt this
    /// object, so care must be taken when using this method.
    pub fn get_mut(&mut self) -> &mut W {
        self.inner.get_mut().get_mut()
    }

    /// Attempt to finish this output stream, writing out final chunks of data.
    ///
    /// Note that this function can only be used once data has finished being
    /// written to the output stream. After this function is called then further
    /// calls to `write` may result in a panic.
    ///
    /// # Panics
    ///
    /// Attempts to write data to this stream may result in a panic after this
    /// function is called.
    ///
    /// # Errors
    ///
    /// This function will perform I/O to finish the stream, returning any
    /// errors which happen.
    pub fn try_finish(&mut self) -> io::Result<()> {
        self.finish_and_check_crc()?;
        Ok(())
    }

    /// Consumes this decoder, flushing the output stream.
    ///
    /// This will flush the underlying data stream and then return the contained
    /// writer if the flush succeeded.
    ///
    /// Note that this function may not be suitable to call in a situation where
    /// the underlying stream is an asynchronous I/O stream. To finish a stream
    /// the `try_finish` (or `shutdown`) method should be used instead. To
    /// re-acquire ownership of a stream it is safe to call this method after
    /// `try_finish` or `shutdown` has returned `Ok`.
    ///
    /// # Errors
    ///
    /// This function will perform I/O to complete this stream, and any I/O
    /// errors which occur will be returned from this function.
    pub fn finish(mut self) -> io::Result<W> {
        self.finish_and_check_crc()?;
        Ok(self.inner.take_inner().into_inner())
    }

    fn finish_and_check_crc(&mut self) -> io::Result<()> {
        self.inner.finish()?;

        if self.crc_bytes.len() != 8 {
            return Err(corrupt());
        }

        let crc = ((self.crc_bytes[0] as u32) << 0)
            | ((self.crc_bytes[1] as u32) << 8)
            | ((self.crc_bytes[2] as u32) << 16)
            | ((self.crc_bytes[3] as u32) << 24);
        let amt = ((self.crc_bytes[4] as u32) << 0)
            | ((self.crc_bytes[5] as u32) << 8)
            | ((self.crc_bytes[6] as u32) << 16)
            | ((self.crc_bytes[7] as u32) << 24);
        if crc != self.inner.get_ref().crc().sum() {
            return Err(corrupt());
        }
        if amt != self.inner.get_ref().crc().amount() {
            return Err(corrupt());
        }
        Ok(())
    }
}

impl<W: Write> Write for GzDecoder<W> {
    fn write(&mut self, mut buf: &[u8]) -> io::Result<usize> {
        let buflen = buf.len();
        if self.header().is_none() {
            match self.header_parser.parse(&mut buf) {
                Err(err) => {
                    if err.kind() == io::ErrorKind::UnexpectedEof {
                        // all data read but header still not complete
                        Ok(buflen)
                    } else {
                        Err(err)
                    }
                }
                Ok(_) => {
                    debug_assert!(self.header().is_some());
                    // buf now contains the unread part of the original buf
                    let n = buflen - buf.len();
                    Ok(n)
                }
            }
        } else {
            let (n, status) = self.inner.write_with_status(buf)?;

            if status == Status::StreamEnd && n < buf.len() && self.crc_bytes.len() < 8 {
                let remaining = buf.len() - n;
                let crc_bytes = cmp::min(remaining, CRC_BYTES_LEN - self.crc_bytes.len());
                self.crc_bytes.extend(&buf[n..n + crc_bytes]);
                return Ok(n + crc_bytes);
            }
            Ok(n)
        }
    }

    fn flush(&mut self) -> io::Result<()> {
        self.inner.flush()
    }
}

impl<W: Read + Write> Read for GzDecoder<W> {
    fn read(&mut self, buf: &mut [u8]) -> io::Result<usize> {
        self.inner.get_mut().get_mut().read(buf)
    }
}

/// A gzip streaming decoder that decodes a [gzip file] with multiple members.
///
/// This structure exposes a [`Write`] interface that will consume compressed data and
/// write uncompressed data to the underlying writer.
///
/// A gzip file consists of a series of *members* concatenated one after another.
/// `MultiGzDecoder` decodes all members of a file and writes them to the
/// underlying writer one after another.
///
/// To handle members separately, see [GzDecoder] or read more
/// [in the introduction](../index.html#about-multi-member-gzip-files).
///
/// [gzip file]: https://www.rfc-editor.org/rfc/rfc1952#page-5
#[derive(Debug)]
pub struct MultiGzDecoder<W: Write> {
    inner: GzDecoder<W>,
}

impl<W: Write> MultiGzDecoder<W> {
    /// Creates a new decoder which will write uncompressed data to the stream.
    /// If the gzip stream contains multiple members all will be decoded.
    pub fn new(w: W) -> MultiGzDecoder<W> {
        MultiGzDecoder {
            inner: GzDecoder::new(w),
        }
    }

    /// Returns the header associated with the current member.
    pub fn header(&self) -> Option<&GzHeader> {
        self.inner.header()
    }

    /// Acquires a reference to the underlying writer.
    pub fn get_ref(&self) -> &W {
        self.inner.get_ref()
    }

    /// Acquires a mutable reference to the underlying writer.
    ///
    /// Note that mutating the output/input state of the stream may corrupt this
    /// object, so care must be taken when using this method.
    pub fn get_mut(&mut self) -> &mut W {
        self.inner.get_mut()
    }

    /// Attempt to finish this output stream, writing out final chunks of data.
    ///
    /// Note that this function can only be used once data has finished being
    /// written to the output stream. After this function is called then further
    /// calls to `write` may result in a panic.
    ///
    /// # Panics
    ///
    /// Attempts to write data to this stream may result in a panic after this
    /// function is called.
    ///
    /// # Errors
    ///
    /// This function will perform I/O to finish the stream, returning any
    /// errors which happen.
    pub fn try_finish(&mut self) -> io::Result<()> {
        self.inner.try_finish()
    }

    /// Consumes this decoder, flushing the output stream.
    ///
    /// This will flush the underlying data stream and then return the contained
    /// writer if the flush succeeded.
    ///
    /// Note that this function may not be suitable to call in a situation where
    /// the underlying stream is an asynchronous I/O stream. To finish a stream
    /// the `try_finish` (or `shutdown`) method should be used instead. To
    /// re-acquire ownership of a stream it is safe to call this method after
    /// `try_finish` or `shutdown` has returned `Ok`.
    ///
    /// # Errors
    ///
    /// This function will perform I/O to complete this stream, and any I/O
    /// errors which occur will be returned from this function.
    pub fn finish(self) -> io::Result<W> {
        self.inner.finish()
    }
}

impl<W: Write> Write for MultiGzDecoder<W> {
    fn write(&mut self, buf: &[u8]) -> io::Result<usize> {
        if buf.is_empty() {
            Ok(0)
        } else {
            match self.inner.write(buf) {
                Ok(0) => {
                    // When the GzDecoder indicates that it has finished
                    // create a new GzDecoder to handle additional data.
                    self.inner.try_finish()?;
                    let w = self.inner.inner.take_inner().into_inner();
                    self.inner = GzDecoder::new(w);
                    self.inner.write(buf)
                }
                res => res,
            }
        }
    }

    fn flush(&mut self) -> io::Result<()> {
        self.inner.flush()
    }
}

#[cfg(test)]
mod tests {
    use super::*;

    const STR: &str = "Hello World Hello World Hello World Hello World Hello World \
                               Hello World Hello World Hello World Hello World Hello World \
                               Hello World Hello World Hello World Hello World Hello World \
                               Hello World Hello World Hello World Hello World Hello World \
                               Hello World Hello World Hello World Hello World Hello World";

    #[test]
    fn decode_writer_one_chunk() {
        let mut e = GzEncoder::new(Vec::new(), Compression::default());
        e.write(STR.as_ref()).unwrap();
        let bytes = e.finish().unwrap();

        let mut writer = Vec::new();
        let mut decoder = GzDecoder::new(writer);
        let n = decoder.write(&bytes[..]).unwrap();
        decoder.write(&bytes[n..]).unwrap();
        decoder.try_finish().unwrap();
        writer = decoder.finish().unwrap();
        let return_string = String::from_utf8(writer).expect("String parsing error");
        assert_eq!(return_string, STR);
    }

    #[test]
    fn decode_writer_partial_header() {
        let mut e = GzEncoder::new(Vec::new(), Compression::default());
        e.write(STR.as_ref()).unwrap();
        let bytes = e.finish().unwrap();

        let mut writer = Vec::new();
        let mut decoder = GzDecoder::new(writer);
        assert_eq!(decoder.write(&bytes[..5]).unwrap(), 5);
        let n = decoder.write(&bytes[5..]).unwrap();
        if n < bytes.len() - 5 {
            decoder.write(&bytes[n + 5..]).unwrap();
        }
        writer = decoder.finish().unwrap();
        let return_string = String::from_utf8(writer).expect("String parsing error");
        assert_eq!(return_string, STR);
    }

    #[test]
    fn decode_writer_partial_header_filename() {
        let filename = "test.txt";
        let mut e = GzBuilder::new()
            .filename(filename)
            .read(STR.as_bytes(), Compression::default());
        let mut bytes = Vec::new();
        e.read_to_end(&mut bytes).unwrap();

        let mut writer = Vec::new();
        let mut decoder = GzDecoder::new(writer);
        assert_eq!(decoder.write(&bytes[..12]).unwrap(), 12);
        let n = decoder.write(&bytes[12..]).unwrap();
        if n < bytes.len() - 12 {
            decoder.write(&bytes[n + 12..]).unwrap();
        }
        assert_eq!(
            decoder.header().unwrap().filename().unwrap(),
            filename.as_bytes()
        );
        writer = decoder.finish().unwrap();
        let return_string = String::from_utf8(writer).expect("String parsing error");
        assert_eq!(return_string, STR);
    }

    #[test]
    fn decode_writer_partial_header_comment() {
        let comment = "test comment";
        let mut e = GzBuilder::new()
            .comment(comment)
            .read(STR.as_bytes(), Compression::default());
        let mut bytes = Vec::new();
        e.read_to_end(&mut bytes).unwrap();

        let mut writer = Vec::new();
        let mut decoder = GzDecoder::new(writer);
        assert_eq!(decoder.write(&bytes[..12]).unwrap(), 12);
        let n = decoder.write(&bytes[12..]).unwrap();
        if n < bytes.len() - 12 {
            decoder.write(&bytes[n + 12..]).unwrap();
        }
        assert_eq!(
            decoder.header().unwrap().comment().unwrap(),
            comment.as_bytes()
        );
        writer = decoder.finish().unwrap();
        let return_string = String::from_utf8(writer).expect("String parsing error");
        assert_eq!(return_string, STR);
    }

    #[test]
    fn decode_writer_exact_header() {
        let mut e = GzEncoder::new(Vec::new(), Compression::default());
        e.write(STR.as_ref()).unwrap();
        let bytes = e.finish().unwrap();

        let mut writer = Vec::new();
        let mut decoder = GzDecoder::new(writer);
        assert_eq!(decoder.write(&bytes[..10]).unwrap(), 10);
        decoder.write(&bytes[10..]).unwrap();
        writer = decoder.finish().unwrap();
        let return_string = String::from_utf8(writer).expect("String parsing error");
        assert_eq!(return_string, STR);
    }

    #[test]
    fn decode_writer_partial_crc() {
        let mut e = GzEncoder::new(Vec::new(), Compression::default());
        e.write(STR.as_ref()).unwrap();
        let bytes = e.finish().unwrap();

        let mut writer = Vec::new();
        let mut decoder = GzDecoder::new(writer);
        let l = bytes.len() - 5;
        let n = decoder.write(&bytes[..l]).unwrap();
        decoder.write(&bytes[n..]).unwrap();
        writer = decoder.finish().unwrap();
        let return_string = String::from_utf8(writer).expect("String parsing error");
        assert_eq!(return_string, STR);
    }

    // Two or more gzip files concatenated form a multi-member gzip file. MultiGzDecoder will
    // concatenate the decoded contents of all members.
    #[test]
    fn decode_multi_writer() {
        let mut e = GzEncoder::new(Vec::new(), Compression::default());
        e.write(STR.as_ref()).unwrap();
        let bytes = e.finish().unwrap().repeat(2);

        let mut writer = Vec::new();
        let mut decoder = MultiGzDecoder::new(writer);
        let mut count = 0;
        while count < bytes.len() {
            let n = decoder.write(&bytes[count..]).unwrap();
            assert!(n != 0);
            count += n;
        }
        writer = decoder.finish().unwrap();
        let return_string = String::from_utf8(writer).expect("String parsing error");
        let expected = STR.repeat(2);
        assert_eq!(return_string, expected);
    }

    // GzDecoder consumes one gzip member and then returns 0 for subsequent writes, allowing any
    // additional data to be consumed by the caller.
    #[test]
    fn decode_extra_data() {
        let compressed = {
            let mut e = GzEncoder::new(Vec::new(), Compression::default());
            e.write(STR.as_ref()).unwrap();
            let mut b = e.finish().unwrap();
            b.push(b'x');
            b
        };

        let mut writer = Vec::new();
        let mut decoder = GzDecoder::new(writer);
        let mut consumed_bytes = 0;
        loop {
            let n = decoder.write(&compressed[consumed_bytes..]).unwrap();
            if n == 0 {
                break;
            }
            consumed_bytes += n;
        }
        writer = decoder.finish().unwrap();
        let actual = String::from_utf8(writer).expect("String parsing error");
        assert_eq!(actual, STR);
        assert_eq!(&compressed[consumed_bytes..], b"x");
    }
}