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
|
import io
import os
import random
import struct
import unittest
import zstandard as zstd
from .common import (
CustomBytesIO,
)
class TestDecompressor_read_to_iter(unittest.TestCase):
def test_type_validation(self):
dctx = zstd.ZstdDecompressor()
# Object with read() works.
dctx.read_to_iter(io.BytesIO())
# Buffer protocol works.
dctx.read_to_iter(b"foobar")
with self.assertRaisesRegex(
ValueError, "must pass an object with a read"
):
b"".join(dctx.read_to_iter(True))
def test_empty_input(self):
dctx = zstd.ZstdDecompressor()
source = io.BytesIO()
it = dctx.read_to_iter(source)
# TODO this is arguably wrong. Should get an error about missing frame foo.
with self.assertRaises(StopIteration):
next(it)
it = dctx.read_to_iter(b"")
with self.assertRaises(StopIteration):
next(it)
def test_invalid_input(self):
dctx = zstd.ZstdDecompressor()
source = io.BytesIO(b"foobar")
it = dctx.read_to_iter(source)
with self.assertRaisesRegex(zstd.ZstdError, "Unknown frame descriptor"):
next(it)
it = dctx.read_to_iter(b"foobar")
with self.assertRaisesRegex(zstd.ZstdError, "Unknown frame descriptor"):
next(it)
def test_empty_roundtrip(self):
cctx = zstd.ZstdCompressor(level=1, write_content_size=False)
empty = cctx.compress(b"")
source = io.BytesIO(empty)
source.seek(0)
dctx = zstd.ZstdDecompressor()
it = dctx.read_to_iter(source)
# No chunks should be emitted since there is no data.
with self.assertRaises(StopIteration):
next(it)
# Again for good measure.
with self.assertRaises(StopIteration):
next(it)
def test_skip_bytes_too_large(self):
dctx = zstd.ZstdDecompressor()
with self.assertRaisesRegex(
ValueError, "skip_bytes must be smaller than read_size"
):
b"".join(dctx.read_to_iter(b"", skip_bytes=1, read_size=1))
with self.assertRaisesRegex(
ValueError, "skip_bytes larger than first input chunk"
):
b"".join(dctx.read_to_iter(b"foobar", skip_bytes=10))
def test_skip_bytes(self):
cctx = zstd.ZstdCompressor(write_content_size=False)
compressed = cctx.compress(b"foobar")
dctx = zstd.ZstdDecompressor()
output = b"".join(dctx.read_to_iter(b"hdr" + compressed, skip_bytes=3))
self.assertEqual(output, b"foobar")
def test_large_output(self):
source = io.BytesIO()
source.write(b"f" * zstd.DECOMPRESSION_RECOMMENDED_OUTPUT_SIZE)
source.write(b"o")
source.seek(0)
cctx = zstd.ZstdCompressor(level=1)
compressed = io.BytesIO(cctx.compress(source.getvalue()))
compressed.seek(0)
dctx = zstd.ZstdDecompressor()
it = dctx.read_to_iter(compressed)
chunks = []
chunks.append(next(it))
chunks.append(next(it))
with self.assertRaises(StopIteration):
next(it)
decompressed = b"".join(chunks)
self.assertEqual(decompressed, source.getvalue())
# And again with buffer protocol.
it = dctx.read_to_iter(compressed.getvalue())
chunks = []
chunks.append(next(it))
chunks.append(next(it))
with self.assertRaises(StopIteration):
next(it)
decompressed = b"".join(chunks)
self.assertEqual(decompressed, source.getvalue())
@unittest.skipUnless(
"ZSTD_SLOW_TESTS" in os.environ, "ZSTD_SLOW_TESTS not set"
)
def test_large_input(self):
bytes = list(struct.Struct(">B").pack(i) for i in range(256))
compressed = io.BytesIO()
input_size = 0
cctx = zstd.ZstdCompressor(level=1)
with cctx.stream_writer(compressed, closefd=False) as compressor:
while True:
compressor.write(random.choice(bytes))
input_size += 1
have_compressed = (
len(compressed.getvalue())
> zstd.DECOMPRESSION_RECOMMENDED_INPUT_SIZE
)
have_raw = (
input_size > zstd.DECOMPRESSION_RECOMMENDED_OUTPUT_SIZE * 2
)
if have_compressed and have_raw:
break
compressed = io.BytesIO(compressed.getvalue())
self.assertGreater(
len(compressed.getvalue()),
zstd.DECOMPRESSION_RECOMMENDED_INPUT_SIZE,
)
dctx = zstd.ZstdDecompressor()
it = dctx.read_to_iter(compressed)
chunks = []
chunks.append(next(it))
chunks.append(next(it))
chunks.append(next(it))
with self.assertRaises(StopIteration):
next(it)
decompressed = b"".join(chunks)
self.assertEqual(len(decompressed), input_size)
# And again with buffer protocol.
it = dctx.read_to_iter(compressed.getvalue())
chunks = []
chunks.append(next(it))
chunks.append(next(it))
chunks.append(next(it))
with self.assertRaises(StopIteration):
next(it)
decompressed = b"".join(chunks)
self.assertEqual(len(decompressed), input_size)
def test_interesting(self):
# Found this edge case via fuzzing.
cctx = zstd.ZstdCompressor(level=1)
source = io.BytesIO()
compressed = io.BytesIO()
with cctx.stream_writer(compressed, closefd=False) as compressor:
for i in range(256):
chunk = b"\0" * 1024
compressor.write(chunk)
source.write(chunk)
dctx = zstd.ZstdDecompressor()
simple = dctx.decompress(
compressed.getvalue(), max_output_size=len(source.getvalue())
)
self.assertEqual(simple, source.getvalue())
compressed = io.BytesIO(compressed.getvalue())
streamed = b"".join(dctx.read_to_iter(compressed))
self.assertEqual(streamed, source.getvalue())
def test_read_write_size(self):
source = CustomBytesIO(zstd.ZstdCompressor().compress(b"foobarfoobar"))
dctx = zstd.ZstdDecompressor()
for chunk in dctx.read_to_iter(source, read_size=1, write_size=1):
self.assertEqual(len(chunk), 1)
self.assertEqual(source._read_count, len(source.getvalue()))
def test_magic_less(self):
params = zstd.ZstdCompressionParameters.from_level(
1, format=zstd.FORMAT_ZSTD1_MAGICLESS
)
cctx = zstd.ZstdCompressor(compression_params=params)
frame = cctx.compress(b"foobar")
self.assertNotEqual(frame[0:4], b"\x28\xb5\x2f\xfd")
dctx = zstd.ZstdDecompressor()
with self.assertRaisesRegex(
zstd.ZstdError, "error determining content size from frame header"
):
dctx.decompress(frame)
dctx = zstd.ZstdDecompressor(format=zstd.FORMAT_ZSTD1_MAGICLESS)
res = b"".join(dctx.read_to_iter(frame))
self.assertEqual(res, b"foobar")
|
