forked from RustPython/RustPython
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathzlib.rs
More file actions
590 lines (534 loc) · 19.8 KB
/
zlib.rs
File metadata and controls
590 lines (534 loc) · 19.8 KB
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
// spell-checker:ignore compressobj decompressobj zdict chunksize zlibmodule miniz
pub(crate) use zlib::make_module;
#[pymodule]
mod zlib {
use crate::vm::{
builtins::{PyBaseExceptionRef, PyBytes, PyBytesRef, PyIntRef, PyTypeRef},
common::lock::PyMutex,
convert::TryFromBorrowedObject,
function::{ArgBytesLike, ArgPrimitiveIndex, ArgSize, OptionalArg},
PyObject, PyPayload, PyResult, VirtualMachine,
};
use adler32::RollingAdler32 as Adler32;
use crossbeam_utils::atomic::AtomicCell;
use flate2::{
write::ZlibEncoder, Compress, Compression, Decompress, FlushCompress, FlushDecompress,
Status,
};
use std::io::Write;
#[cfg(not(feature = "zlib"))]
mod constants {
pub const Z_NO_COMPRESSION: i32 = 0;
pub const Z_BEST_COMPRESSION: i32 = 9;
pub const Z_BEST_SPEED: i32 = 1;
pub const Z_DEFAULT_COMPRESSION: i32 = -1;
pub const Z_NO_FLUSH: i32 = 0;
pub const Z_PARTIAL_FLUSH: i32 = 1;
pub const Z_SYNC_FLUSH: i32 = 2;
pub const Z_FULL_FLUSH: i32 = 3;
// not sure what the value here means, but it's the only compression method zlibmodule
// supports, so it doesn't really matter
pub const Z_DEFLATED: i32 = 8;
}
#[cfg(feature = "zlib")]
use libz_sys as constants;
#[pyattr]
use constants::{
Z_BEST_COMPRESSION, Z_BEST_SPEED, Z_DEFAULT_COMPRESSION, Z_DEFLATED as DEFLATED,
Z_FULL_FLUSH, Z_NO_COMPRESSION, Z_NO_FLUSH, Z_PARTIAL_FLUSH, Z_SYNC_FLUSH,
};
#[cfg(feature = "zlib")]
#[pyattr]
use libz_sys::{
Z_BLOCK, Z_DEFAULT_STRATEGY, Z_FILTERED, Z_FINISH, Z_FIXED, Z_HUFFMAN_ONLY, Z_RLE, Z_TREES,
};
// copied from zlibmodule.c (commit 530f506ac91338)
#[pyattr]
const MAX_WBITS: i8 = 15;
#[pyattr]
const DEF_BUF_SIZE: usize = 16 * 1024;
#[pyattr]
const DEF_MEM_LEVEL: u8 = 8;
#[pyattr(once)]
fn error(vm: &VirtualMachine) -> PyTypeRef {
vm.ctx.new_exception_type(
"zlib",
"error",
Some(vec![vm.ctx.exceptions.exception_type.to_owned()]),
)
}
#[pyfunction]
fn adler32(data: ArgBytesLike, begin_state: OptionalArg<PyIntRef>) -> u32 {
data.with_ref(|data| {
let begin_state = begin_state.map_or(1, |i| i.as_u32_mask());
let mut hasher = Adler32::from_value(begin_state);
hasher.update_buffer(data);
hasher.hash()
})
}
#[pyfunction]
fn crc32(data: ArgBytesLike, begin_state: OptionalArg<PyIntRef>) -> u32 {
crate::binascii::crc32(data, begin_state)
}
#[derive(FromArgs)]
struct PyFuncCompressArgs {
#[pyarg(positional)]
data: ArgBytesLike,
#[pyarg(any, default = "Level::new(Z_DEFAULT_COMPRESSION)")]
level: Level,
#[pyarg(any, default = "ArgPrimitiveIndex { value: MAX_WBITS }")]
wbits: ArgPrimitiveIndex<i8>,
}
/// Returns a bytes object containing compressed data.
#[pyfunction]
fn compress(args: PyFuncCompressArgs, vm: &VirtualMachine) -> PyResult<PyBytesRef> {
let PyFuncCompressArgs {
data,
level,
ref wbits,
} = args;
let level = level.ok_or_else(|| new_zlib_error("Bad compression level", vm))?;
let encoded_bytes = if args.wbits.value == MAX_WBITS {
let mut encoder = ZlibEncoder::new(Vec::new(), level);
data.with_ref(|input_bytes| encoder.write_all(input_bytes).unwrap());
encoder.finish().unwrap()
} else {
let mut inner = CompressInner::new(InitOptions::new(wbits.value, vm)?.compress(level));
data.with_ref(|input_bytes| inner.compress(input_bytes, vm))?;
inner.flush(vm)?
};
Ok(vm.ctx.new_bytes(encoded_bytes))
}
enum InitOptions {
Standard {
header: bool,
// [De]Compress::new_with_window_bits is only enabled for zlib; miniz_oxide doesn't
// support wbits (yet?)
#[cfg(feature = "zlib")]
wbits: u8,
},
#[cfg(feature = "zlib")]
Gzip { wbits: u8 },
}
impl InitOptions {
fn new(wbits: i8, vm: &VirtualMachine) -> PyResult<InitOptions> {
let header = wbits > 0;
let wbits = wbits.unsigned_abs();
match wbits {
9..=15 => Ok(InitOptions::Standard {
header,
#[cfg(feature = "zlib")]
wbits,
}),
#[cfg(feature = "zlib")]
25..=31 => Ok(InitOptions::Gzip { wbits: wbits - 16 }),
_ => Err(vm.new_value_error("Invalid initialization option".to_owned())),
}
}
fn decompress(self) -> Decompress {
match self {
#[cfg(not(feature = "zlib"))]
Self::Standard { header } => Decompress::new(header),
#[cfg(feature = "zlib")]
Self::Standard { header, wbits } => Decompress::new_with_window_bits(header, wbits),
#[cfg(feature = "zlib")]
Self::Gzip { wbits } => Decompress::new_gzip(wbits),
}
}
fn compress(self, level: Compression) -> Compress {
match self {
#[cfg(not(feature = "zlib"))]
Self::Standard { header } => Compress::new(level, header),
#[cfg(feature = "zlib")]
Self::Standard { header, wbits } => {
Compress::new_with_window_bits(level, header, wbits)
}
#[cfg(feature = "zlib")]
Self::Gzip { wbits } => Compress::new_gzip(level, wbits),
}
}
}
fn _decompress(
mut data: &[u8],
d: &mut Decompress,
bufsize: usize,
max_length: Option<usize>,
is_flush: bool,
vm: &VirtualMachine,
) -> PyResult<(Vec<u8>, bool)> {
if data.is_empty() {
return Ok((Vec::new(), true));
}
let mut buf = Vec::new();
loop {
let final_chunk = data.len() <= CHUNKSIZE;
let chunk = if final_chunk {
data
} else {
&data[..CHUNKSIZE]
};
// if this is the final chunk, finish it
let flush = if is_flush {
if final_chunk {
FlushDecompress::Finish
} else {
FlushDecompress::None
}
} else {
FlushDecompress::Sync
};
loop {
let additional = if let Some(max_length) = max_length {
std::cmp::min(bufsize, max_length - buf.capacity())
} else {
bufsize
};
if additional == 0 {
return Ok((buf, false));
}
buf.reserve_exact(additional);
let prev_in = d.total_in();
let status = d
.decompress_vec(chunk, &mut buf, flush)
.map_err(|_| new_zlib_error("invalid input data", vm))?;
let consumed = d.total_in() - prev_in;
data = &data[consumed as usize..];
let stream_end = status == Status::StreamEnd;
if stream_end || data.is_empty() {
// we've reached the end of the stream, we're done
buf.shrink_to_fit();
return Ok((buf, stream_end));
} else if !chunk.is_empty() && consumed == 0 {
// we're gonna need a bigger buffer
continue;
} else {
// next chunk
break;
8000
}
}
}
}
#[derive(FromArgs)]
struct PyFuncDecompressArgs {
#[pyarg(positional)]
data: ArgBytesLike,
#[pyarg(any, default = "ArgPrimitiveIndex { value: MAX_WBITS }")]
wbits: ArgPrimitiveIndex<i8>,
#[pyarg(any, default = "ArgPrimitiveIndex { value: DEF_BUF_SIZE }")]
bufsize: ArgPrimitiveIndex<usize>,
}
/// Returns a bytes object containing the uncompressed data.
#[pyfunction]
fn decompress(args: PyFuncDecompressArgs, vm: &VirtualMachine) -> PyResult<Vec<u8>> {
let PyFuncDecompressArgs {
data,
wbits,
bufsize,
} = args;
data.with_ref(|data| {
let mut d = InitOptions::new(wbits.value, vm)?.decompress();
let (buf, stream_end) = _decompress(data, &mut d, bufsize.value, None, false, vm)?;
if !stream_end {
return Err(new_zlib_error(
"Error -5 while decompressing data: incomplete or truncated stream",
vm,
));
}
Ok(buf)
})
}
#[derive(FromArgs)]
struct DecompressobjArgs {
#[pyarg(any, default = "ArgPrimitiveIndex { value: MAX_WBITS }")]
wbits: ArgPrimitiveIndex<i8>,
#[cfg(feature = "zlib")]
#[pyarg(any, optional)]
_zdict: OptionalArg<ArgBytesLike>,
}
#[pyfunction]
fn decompressobj(args: DecompressobjArgs, vm: &VirtualMachine) -> PyResult<PyDecompress> {
#[allow(unused_mut)]
let mut decompress = InitOptions::new(args.wbits.value, vm)?.decompress();
#[cfg(feature = "zlib")]
if let OptionalArg::Present(_dict) = args._zdict {
// FIXME: always fails
// dict.with_ref(|d| decompress.set_dictionary(d));
}
Ok(PyDecompress {
decompress: PyMutex::new(decompress),
eof: AtomicCell::new(false),
unused_data: PyMutex::new(PyBytes::from(vec![]).into_ref(vm)),
unconsumed_tail: PyMutex::new(PyBytes::from(vec![]).into_ref(vm)),
})
}
#[pyattr]
#[pyclass(name = "Decompress")]
#[derive(Debug, PyPayload)]
struct PyDecompress {
decompress: PyMutex<Decompress>,
eof: AtomicCell<bool>,
unused_data: PyMutex<PyBytesRef>,
unconsumed_tail: PyMutex<PyBytesRef>,
}
#[pyclass]
impl PyDecompress {
#[pygetset]
fn eof(&self) -> bool {
self.eof.load()
}
#[pygetset]
fn unused_data(&self) -> PyBytesRef {
self.unused_data.lock().clone()
}
#[pygetset]
fn unconsumed_tail(&self) -> PyBytesRef {
self.unconsumed_tail.lock().clone()
}
fn save_unused_input(
&self,
d: &mut Decompress,
data: &[u8],
stream_end: bool,
orig_in: u64,
vm: &VirtualMachine,
) {
let leftover = &data[(d.total_in() - orig_in) as usize..];
if stream_end && !leftover.is_empty() {
let mut unused_data = self.unused_data.lock();
let unused: Vec<_> = unused_data
.as_bytes()
.iter()
.chain(leftover)
.copied()
.collect();
*unused_data = vm.new_pyref(unused);
}
}
#[pymethod]
fn decompress(&self, args: DecompressArgs, vm: &VirtualMachine) -> PyResult<Vec<u8>> {
let max_length = args.max_length.value;
let max_length = (max_length != 0).then_some(max_length);
let data = args.data.borrow_buf();
let data = &*data;
let mut d = self.decompress.lock();
let orig_in = d.total_in();
let (ret, stream_end) =
match _decompress(data, &mut d, DEF_BUF_SIZE, max_length, false, vm) {
Ok((buf, true)) => {
self.eof.store(true);
(Ok(buf), true)
}
Ok((buf, false)) => (Ok(buf), false),
Err(err) => (Err(err), false),
};
self.save_unused_input(&mut d, data, stream_end, orig_in, vm);
let leftover = if stream_end {
b""
} else {
&data[(d.total_in() - orig_in) as usize..]
};
let mut unconsumed_tail = self.unconsumed_tail.lock();
if !leftover.is_empty() || !unconsumed_tail.is_empty() {
*unconsumed_tail = PyBytes::from(leftover.to_owned()).into_ref(vm);
}
ret
}
#[pymethod]
fn flush(&self, length: OptionalArg<ArgSize>, vm: &VirtualMachine) -> PyResult<Vec<u8>> {
let length = match length {
OptionalArg::Present(l) => {
let l: isize = l.into();
if l <= 0 {
return Err(
vm.new_value_error("length must be greater than zero".to_owned())
);
} else {
l as usize
}
}
OptionalArg::Missing => DEF_BUF_SIZE,
};
let mut data = self.unconsumed_tail.lock();
let mut d = self.decompress.lock();
let orig_in = d.total_in();
let (ret, stream_end) = match _decompress(&data, &mut d, length, None, true, vm) {
Ok((buf, stream_end)) => (Ok(buf), stream_end),
Err(err) => (Err(err), false),
};
self.save_unused_input(&mut d, &data, stream_end, orig_in, vm);
*data = PyBytes::from(Vec::new()).into_ref(vm);
// TODO: drop the inner decompressor, somehow
// if stream_end {
//
// }
ret
}
}
#[derive(FromArgs)]
struct DecompressArgs {
#[pyarg(positional)]
data: ArgBytesLike,
#[pyarg(
any,
default = "rustpython_vm::function::ArgPrimitiveIndex { value: 0 }"
)]
max_length: ArgPrimitiveIndex<usize>,
}
#[derive(FromArgs)]
#[allow(dead_code)] // FIXME: use args
struct CompressobjArgs {
#[pyarg(any, default = "Level::new(Z_DEFAULT_COMPRESSION)")]
level: Level,
// only DEFLATED is valid right now, it's w/e
#[pyarg(any, default = "DEFLATED")]
_method: i32,
#[pyarg(any, default = "ArgPrimitiveIndex { value: MAX_WBITS }")]
wbits: ArgPrimitiveIndex<i8>,
#[pyarg(any, name = "_memLevel", default = "DEF_MEM_LEVEL")]
_mem_level: u8,
#[cfg(feature = "zlib")]
#[pyarg(any, default = "Z_DEFAULT_STRATEGY")]
_strategy: i32,
#[cfg(feature = "zlib")]
#[pyarg(any, optional)]
zdict: Option<ArgBytesLike>,
}
#[pyfunction]
fn compressobj(args: CompressobjArgs, vm: &VirtualMachine) -> PyResult<PyCompress> {
let CompressobjArgs {
level,
wbits,
#[cfg(feature = "zlib")]
zdict,
..
} = args;
let level =
level.ok_or_else(|| vm.new_value_error("invalid initialization option".to_owned()))?;
#[allow(unused_mut)]
let mut compress = InitOptions::new(wbits.value, vm)?.compress(level);
#[cfg(feature = "zlib")]
if let Some(zdict) = zdict {
zdict.with_ref(|zdict| compress.set_dictionary(zdict).unwrap());
}
Ok(PyCompress {
inner: PyMutex::new(CompressInner::new(compress)),
})
}
#[derive(Debug)]
struct CompressInner {
compress: Compress,
unconsumed: Vec<u8>,
}
#[pyattr]
#[pyclass(name = "Compress")]
#[derive(Debug, PyPayload)]
struct PyCompress {
inner: PyMutex<CompressInner>,
}
#[pyclass]
impl PyCompress {
#[pymethod]
fn compress(&self, data: ArgBytesLike, vm: &VirtualMachine) -> PyResult<Vec<u8>> {
let mut inner = self.inner.lock();
data.with_ref(|b| inner.compress(b, vm))
}
// TODO: mode argument isn't used
#[pymethod]
fn flush(&self, _mode: OptionalArg<i32>, vm: &VirtualMachine) -> PyResult<Vec<u8>> {
self.inner.lock().flush(vm)
}
// TODO: This is an optional feature of Compress
// #[pymethod]
// #[pymethod(magic)]
// #[pymethod(name = "__deepcopy__")]
// fn copy(&self) -> Self {
// todo!("<flate2::Compress as Clone>")
// }
}
const CHUNKSIZE: usize = u32::MAX as usize;
impl CompressInner {
fn new(compress: Compress) -> Self {
Self {
compress,
unconsumed: Vec::new(),
}
}
fn compress(&mut self, data: &[u8], vm: &VirtualMachine) -> PyResult<Vec<u8>> {
let orig_in = self.compress.total_in() as usize;
let mut cur_in = 0;
let unconsumed = std::mem::take(&mut self.unconsumed);
let mut buf = Vec::new();
'outer: for chunk in unconsumed.chunks(CHUNKSIZE).chain(data.chunks(CHUNKSIZE)) {
while cur_in < chunk.len() {
buf.reserve(DEF_BUF_SIZE);
let status = self
.compress
.compress_vec(&chunk[cur_in..], &mut buf, FlushCompress::None)
.map_err(|_| {
self.unconsumed.extend_from_slice(&data[cur_in..]);
new_zlib_error("error while compressing", vm)
})?;
cur_in = (self.compress.total_in() as usize) - orig_in;
match status {
Status::Ok => continue,
Status::StreamEnd => break 'outer,
_ => break,
}
}
}
self.unconsumed.extend_from_slice(&data[cur_in..]);
buf.shrink_to_fit();
Ok(buf)
}
// TODO: flush mode (FlushDecompress) parameter
fn flush(&mut self, vm: &VirtualMachine) -> PyResult<Vec<u8>> {
let data = std::mem::take(&mut self.unconsumed);
let mut data_it = data.chunks(CHUNKSIZE);
let mut buf = Vec::new();
loop {
let chunk = data_it.next().unwrap_or(&[]);
if buf.len() == buf.capacity() {
buf.reserve(DEF_BUF_SIZE);
}
let status = self
.compress
.compress_vec(chunk, &mut buf, FlushCompress::Finish)
.map_err(|_| new_zlib_error("error while compressing", vm))?;
match status {
Status::StreamEnd => break,
_ => continue,
}
}
buf.shrink_to_fit();
Ok(buf)
}
}
fn new_zlib_error(message: &str, vm: &VirtualMachine) -> PyBaseExceptionRef {
vm.new_exception_msg(vm.class("zlib", "error"), message.to_owned())
}
struct Level(Option<flate2::Compression>);
impl Level {
fn new(level: i32) -> Self {
let compression = match level {
Z_DEFAULT_COMPRESSION => Compression::default(),
valid_level @ Z_NO_COMPRESSION..=Z_BEST_COMPRESSION => {
Compression::new(valid_level as u32)
}
_ => return Self(None),
};
Self(Some(compression))
}
fn ok_or_else(
self,
f: impl FnOnce() -> PyBaseExceptionRef,
) -> PyResult<flate2::Compression> {
self.0.ok_or_else(f)
}
}
impl<'a> TryFromBorrowedObject<'a> for Level {
fn try_from_borrowed_object(vm: &VirtualMachine, obj: &'a PyObject) -> PyResult<Self> {
let int: i32 = obj.try_index(vm)?.try_to_primitive(vm)?;
Ok(Self::new(int))
}
}
}