| 1 |
"""Test script for the gzip module. |
| 2 |
""" |
| 3 |
|
| 4 |
import array |
| 5 |
import functools |
| 6 |
import gc |
| 7 |
import io |
| 8 |
import os |
| 9 |
import struct |
| 10 |
import sys |
| 11 |
import unittest |
| 12 |
import warnings |
| 13 |
from subprocess import PIPE, Popen |
| 14 |
from test.support import catch_unraisable_exception |
| 15 |
from test.support import import_helper |
| 16 |
from test.support import os_helper |
| 17 |
from test.support import _4G, bigmemtest, requires_subprocess |
| 18 |
from test.support.script_helper import assert_python_ok, assert_python_failure |
| 19 |
|
| 20 |
gzip = import_helper.import_module('gzip') |
| 21 |
zlib = import_helper.import_module('zlib') |
| 22 |
|
| 23 |
data1 = b""" int length=DEFAULTALLOC, err = Z_OK; |
| 24 |
PyObject *RetVal; |
| 25 |
int flushmode = Z_FINISH; |
| 26 |
unsigned long start_total_out; |
| 27 |
|
| 28 |
""" |
| 29 |
|
| 30 |
data2 = b"""/* zlibmodule.c -- gzip-compatible data compression */ |
| 31 |
/* See http://www.gzip.org/zlib/ |
| 32 |
/* See http://www.winimage.com/zLibDll for Windows */ |
| 33 |
""" |
| 34 |
|
| 35 |
|
| 36 |
TEMPDIR = os.path.abspath(os_helper.TESTFN) + '-gzdir' |
| 37 |
|
| 38 |
|
| 39 |
class UnseekableIO(io.BytesIO): |
| 40 |
def seekable(self): |
| 41 |
return False |
| 42 |
|
| 43 |
def tell(self): |
| 44 |
raise io.UnsupportedOperation |
| 45 |
|
| 46 |
def seek(self, *args): |
| 47 |
raise io.UnsupportedOperation |
| 48 |
|
| 49 |
|
| 50 |
class BaseTest(unittest.TestCase): |
| 51 |
filename = os_helper.TESTFN |
| 52 |
|
| 53 |
def setUp(self): |
| 54 |
os_helper.unlink(self.filename) |
| 55 |
|
| 56 |
def tearDown(self): |
| 57 |
os_helper.unlink(self.filename) |
| 58 |
|
| 59 |
|
| 60 |
class TestGzip(BaseTest): |
| 61 |
def write_and_read_back(self, data, mode='b'): |
| 62 |
b_data = bytes(data) |
| 63 |
with gzip.GzipFile(self.filename, 'w'+mode) as f: |
| 64 |
l = f.write(data) |
| 65 |
self.assertEqual(l, len(b_data)) |
| 66 |
with gzip.GzipFile(self.filename, 'r'+mode) as f: |
| 67 |
self.assertEqual(f.read(), b_data) |
| 68 |
|
| 69 |
def test_write(self): |
| 70 |
with gzip.GzipFile(self.filename, 'wb') as f: |
| 71 |
f.write(data1 * 50) |
| 72 |
|
| 73 |
# Try flush and fileno. |
| 74 |
f.flush() |
| 75 |
f.fileno() |
| 76 |
if hasattr(os, 'fsync'): |
| 77 |
os.fsync(f.fileno()) |
| 78 |
f.close() |
| 79 |
|
| 80 |
# Test multiple close() calls. |
| 81 |
f.close() |
| 82 |
|
| 83 |
def test_write_read_with_pathlike_file(self): |
| 84 |
filename = os_helper.FakePath(self.filename) |
| 85 |
with gzip.GzipFile(filename, 'w') as f: |
| 86 |
f.write(data1 * 50) |
| 87 |
self.assertIsInstance(f.name, str) |
| 88 |
self.assertEqual(f.name, self.filename) |
| 89 |
with gzip.GzipFile(filename, 'a') as f: |
| 90 |
f.write(data1) |
| 91 |
with gzip.GzipFile(filename) as f: |
| 92 |
d = f.read() |
| 93 |
self.assertEqual(d, data1 * 51) |
| 94 |
self.assertIsInstance(f.name, str) |
| 95 |
self.assertEqual(f.name, self.filename) |
| 96 |
|
| 97 |
# The following test_write_xy methods test that write accepts |
| 98 |
# the corresponding bytes-like object type as input |
| 99 |
# and that the data written equals bytes(xy) in all cases. |
| 100 |
def test_write_memoryview(self): |
| 101 |
self.write_and_read_back(memoryview(data1 * 50)) |
| 102 |
m = memoryview(bytes(range(256))) |
| 103 |
data = m.cast('B', shape=[8,8,4]) |
| 104 |
self.write_and_read_back(data) |
| 105 |
|
| 106 |
def test_write_bytearray(self): |
| 107 |
self.write_and_read_back(bytearray(data1 * 50)) |
| 108 |
|
| 109 |
def test_write_array(self): |
| 110 |
self.write_and_read_back(array.array('I', data1 * 40)) |
| 111 |
|
| 112 |
def test_write_incompatible_type(self): |
| 113 |
# Test that non-bytes-like types raise TypeError. |
| 114 |
# Issue #21560: attempts to write incompatible types |
| 115 |
# should not affect the state of the fileobject |
| 116 |
with gzip.GzipFile(self.filename, 'wb') as f: |
| 117 |
with self.assertRaises(TypeError): |
| 118 |
f.write('') |
| 119 |
with self.assertRaises(TypeError): |
| 120 |
f.write([]) |
| 121 |
f.write(data1) |
| 122 |
with gzip.GzipFile(self.filename, 'rb') as f: |
| 123 |
self.assertEqual(f.read(), data1) |
| 124 |
|
| 125 |
def test_read(self): |
| 126 |
self.test_write() |
| 127 |
# Try reading. |
| 128 |
with gzip.GzipFile(self.filename, 'r') as f: |
| 129 |
d = f.read() |
| 130 |
self.assertEqual(d, data1*50) |
| 131 |
|
| 132 |
def test_read1(self): |
| 133 |
self.test_write() |
| 134 |
blocks = [] |
| 135 |
nread = 0 |
| 136 |
with gzip.GzipFile(self.filename, 'r') as f: |
| 137 |
while True: |
| 138 |
d = f.read1() |
| 139 |
if not d: |
| 140 |
break |
| 141 |
blocks.append(d) |
| 142 |
nread += len(d) |
| 143 |
# Check that position was updated correctly (see issue10791). |
| 144 |
self.assertEqual(f.tell(), nread) |
| 145 |
self.assertEqual(b''.join(blocks), data1 * 50) |
| 146 |
|
| 147 |
def test_readinto(self): |
| 148 |
# 10MB of uncompressible data to ensure multiple reads |
| 149 |
large_data = os.urandom(10 * 2**20) |
| 150 |
with gzip.GzipFile(self.filename, 'wb') as f: |
| 151 |
f.write(large_data) |
| 152 |
|
| 153 |
buf = bytearray(len(large_data)) |
| 154 |
with gzip.GzipFile(self.filename, 'r') as f: |
| 155 |
nbytes = f.readinto(buf) |
| 156 |
self.assertEqual(nbytes, len(large_data)) |
| 157 |
self.assertEqual(buf, large_data) |
| 158 |
|
| 159 |
def test_readinto1(self): |
| 160 |
# 10MB of uncompressible data to ensure multiple reads |
| 161 |
large_data = os.urandom(10 * 2**20) |
| 162 |
with gzip.GzipFile(self.filename, 'wb') as f: |
| 163 |
f.write(large_data) |
| 164 |
|
| 165 |
nread = 0 |
| 166 |
buf = bytearray(len(large_data)) |
| 167 |
memview = memoryview(buf) # Simplifies slicing |
| 168 |
with gzip.GzipFile(self.filename, 'r') as f: |
| 169 |
for count in range(200): |
| 170 |
nbytes = f.readinto1(memview[nread:]) |
| 171 |
if not nbytes: |
| 172 |
break |
| 173 |
nread += nbytes |
| 174 |
self.assertEqual(f.tell(), nread) |
| 175 |
self.assertEqual(buf, large_data) |
| 176 |
# readinto1() should require multiple loops |
| 177 |
self.assertGreater(count, 1) |
| 178 |
|
| 179 |
@bigmemtest(size=_4G, memuse=1) |
| 180 |
def test_read_large(self, size): |
| 181 |
# Read chunk size over UINT_MAX should be supported, despite zlib's |
| 182 |
# limitation per low-level call |
| 183 |
compressed = gzip.compress(data1, compresslevel=1) |
| 184 |
f = gzip.GzipFile(fileobj=io.BytesIO(compressed), mode='rb') |
| 185 |
self.assertEqual(f.read(size), data1) |
| 186 |
|
| 187 |
def test_io_on_closed_object(self): |
| 188 |
# Test that I/O operations on closed GzipFile objects raise a |
| 189 |
# ValueError, just like the corresponding functions on file objects. |
| 190 |
|
| 191 |
# Write to a file, open it for reading, then close it. |
| 192 |
self.test_write() |
| 193 |
f = gzip.GzipFile(self.filename, 'r') |
| 194 |
fileobj = f.fileobj |
| 195 |
self.assertFalse(fileobj.closed) |
| 196 |
f.close() |
| 197 |
self.assertTrue(fileobj.closed) |
| 198 |
with self.assertRaises(ValueError): |
| 199 |
f.read(1) |
| 200 |
with self.assertRaises(ValueError): |
| 201 |
f.seek(0) |
| 202 |
with self.assertRaises(ValueError): |
| 203 |
f.tell() |
| 204 |
# Open the file for writing, then close it. |
| 205 |
f = gzip.GzipFile(self.filename, 'w') |
| 206 |
fileobj = f.fileobj |
| 207 |
self.assertFalse(fileobj.closed) |
| 208 |
f.close() |
| 209 |
self.assertTrue(fileobj.closed) |
| 210 |
with self.assertRaises(ValueError): |
| 211 |
f.write(b'') |
| 212 |
with self.assertRaises(ValueError): |
| 213 |
f.flush() |
| 214 |
|
| 215 |
def test_append(self): |
| 216 |
self.test_write() |
| 217 |
# Append to the previous file |
| 218 |
with gzip.GzipFile(self.filename, 'ab') as f: |
| 219 |
f.write(data2 * 15) |
| 220 |
|
| 221 |
with gzip.GzipFile(self.filename, 'rb') as f: |
| 222 |
d = f.read() |
| 223 |
self.assertEqual(d, (data1*50) + (data2*15)) |
| 224 |
|
| 225 |
def test_many_append(self): |
| 226 |
# Bug #1074261 was triggered when reading a file that contained |
| 227 |
# many, many members. Create such a file and verify that reading it |
| 228 |
# works. |
| 229 |
with gzip.GzipFile(self.filename, 'wb', 9) as f: |
| 230 |
f.write(b'a') |
| 231 |
for i in range(0, 200): |
| 232 |
with gzip.GzipFile(self.filename, "ab", 9) as f: # append |
| 233 |
f.write(b'a') |
| 234 |
|
| 235 |
# Try reading the file |
| 236 |
with gzip.GzipFile(self.filename, "rb") as zgfile: |
| 237 |
contents = b"" |
| 238 |
while 1: |
| 239 |
ztxt = zgfile.read(8192) |
| 240 |
contents += ztxt |
| 241 |
if not ztxt: break |
| 242 |
self.assertEqual(contents, b'a'*201) |
| 243 |
|
| 244 |
def test_exclusive_write(self): |
| 245 |
with gzip.GzipFile(self.filename, 'xb') as f: |
| 246 |
f.write(data1 * 50) |
| 247 |
with gzip.GzipFile(self.filename, 'rb') as f: |
| 248 |
self.assertEqual(f.read(), data1 * 50) |
| 249 |
with self.assertRaises(FileExistsError): |
| 250 |
gzip.GzipFile(self.filename, 'xb') |
| 251 |
|
| 252 |
def test_buffered_reader(self): |
| 253 |
# Issue #7471: a GzipFile can be wrapped in a BufferedReader for |
| 254 |
# performance. |
| 255 |
self.test_write() |
| 256 |
|
| 257 |
with gzip.GzipFile(self.filename, 'rb') as f: |
| 258 |
with io.BufferedReader(f) as r: |
| 259 |
lines = [line for line in r] |
| 260 |
|
| 261 |
self.assertEqual(lines, 50 * data1.splitlines(keepends=True)) |
| 262 |
|
| 263 |
def test_readline(self): |
| 264 |
self.test_write() |
| 265 |
# Try .readline() with varying line lengths |
| 266 |
|
| 267 |
with gzip.GzipFile(self.filename, 'rb') as f: |
| 268 |
line_length = 0 |
| 269 |
while 1: |
| 270 |
L = f.readline(line_length) |
| 271 |
if not L and line_length != 0: break |
| 272 |
self.assertTrue(len(L) <= line_length) |
| 273 |
line_length = (line_length + 1) % 50 |
| 274 |
|
| 275 |
def test_readlines(self): |
| 276 |
self.test_write() |
| 277 |
# Try .readlines() |
| 278 |
|
| 279 |
with gzip.GzipFile(self.filename, 'rb') as f: |
| 280 |
L = f.readlines() |
| 281 |
|
| 282 |
with gzip.GzipFile(self.filename, 'rb') as f: |
| 283 |
while 1: |
| 284 |
L = f.readlines(150) |
| 285 |
if L == []: break |
| 286 |
|
| 287 |
def test_seek_read(self): |
| 288 |
self.test_write() |
| 289 |
# Try seek, read test |
| 290 |
|
| 291 |
with gzip.GzipFile(self.filename) as f: |
| 292 |
while 1: |
| 293 |
oldpos = f.tell() |
| 294 |
line1 = f.readline() |
| 295 |
if not line1: break |
| 296 |
newpos = f.tell() |
| 297 |
f.seek(oldpos) # negative seek |
| 298 |
if len(line1)>10: |
| 299 |
amount = 10 |
| 300 |
else: |
| 301 |
amount = len(line1) |
| 302 |
line2 = f.read(amount) |
| 303 |
self.assertEqual(line1[:amount], line2) |
| 304 |
f.seek(newpos) # positive seek |
| 305 |
|
| 306 |
def test_seek_whence(self): |
| 307 |
self.test_write() |
| 308 |
# Try seek(whence=1), read test |
| 309 |
|
| 310 |
with gzip.GzipFile(self.filename) as f: |
| 311 |
f.read(10) |
| 312 |
f.seek(10, whence=1) |
| 313 |
y = f.read(10) |
| 314 |
self.assertEqual(y, data1[20:30]) |
| 315 |
|
| 316 |
def test_seek_write(self): |
| 317 |
# Try seek, write test |
| 318 |
with gzip.GzipFile(self.filename, 'w') as f: |
| 319 |
for pos in range(0, 256, 16): |
| 320 |
f.seek(pos) |
| 321 |
f.write(b'GZ\n') |
| 322 |
|
| 323 |
def test_mode(self): |
| 324 |
self.test_write() |
| 325 |
with gzip.GzipFile(self.filename, 'r') as f: |
| 326 |
self.assertEqual(f.myfileobj.mode, 'rb') |
| 327 |
os_helper.unlink(self.filename) |
| 328 |
with gzip.GzipFile(self.filename, 'x') as f: |
| 329 |
self.assertEqual(f.myfileobj.mode, 'xb') |
| 330 |
|
| 331 |
def test_1647484(self): |
| 332 |
for mode in ('wb', 'rb'): |
| 333 |
with gzip.GzipFile(self.filename, mode) as f: |
| 334 |
self.assertHasAttr(f, "name") |
| 335 |
self.assertEqual(f.name, self.filename) |
| 336 |
|
| 337 |
def test_paddedfile_getattr(self): |
| 338 |
self.test_write() |
| 339 |
with gzip.GzipFile(self.filename, 'rb') as f: |
| 340 |
self.assertHasAttr(f.fileobj, "name") |
| 341 |
self.assertEqual(f.fileobj.name, self.filename) |
| 342 |
|
| 343 |
def test_mtime(self): |
| 344 |
mtime = 123456789 |
| 345 |
with gzip.GzipFile(self.filename, 'w', mtime = mtime) as fWrite: |
| 346 |
fWrite.write(data1) |
| 347 |
with gzip.GzipFile(self.filename) as fRead: |
| 348 |
self.assertHasAttr(fRead, 'mtime') |
| 349 |
self.assertIsNone(fRead.mtime) |
| 350 |
dataRead = fRead.read() |
| 351 |
self.assertEqual(dataRead, data1) |
| 352 |
self.assertEqual(fRead.mtime, mtime) |
| 353 |
|
| 354 |
def test_metadata(self): |
| 355 |
mtime = 123456789 |
| 356 |
|
| 357 |
with gzip.GzipFile(self.filename, 'w', mtime = mtime) as fWrite: |
| 358 |
fWrite.write(data1) |
| 359 |
|
| 360 |
with open(self.filename, 'rb') as fRead: |
| 361 |
# see RFC 1952: http://www.faqs.org/rfcs/rfc1952.html |
| 362 |
|
| 363 |
idBytes = fRead.read(2) |
| 364 |
self.assertEqual(idBytes, b'\x1f\x8b') # gzip ID |
| 365 |
|
| 366 |
cmByte = fRead.read(1) |
| 367 |
self.assertEqual(cmByte, b'\x08') # deflate |
| 368 |
|
| 369 |
try: |
| 370 |
expectedname = self.filename.encode('Latin-1') + b'\x00' |
| 371 |
expectedflags = b'\x08' # only the FNAME flag is set |
| 372 |
except UnicodeEncodeError: |
| 373 |
expectedname = b'' |
| 374 |
expectedflags = b'\x00' |
| 375 |
|
| 376 |
flagsByte = fRead.read(1) |
| 377 |
self.assertEqual(flagsByte, expectedflags) |
| 378 |
|
| 379 |
mtimeBytes = fRead.read(4) |
| 380 |
self.assertEqual(mtimeBytes, struct.pack('<i', mtime)) # little-endian |
| 381 |
|
| 382 |
xflByte = fRead.read(1) |
| 383 |
self.assertEqual(xflByte, b'\x02') # maximum compression |
| 384 |
|
| 385 |
osByte = fRead.read(1) |
| 386 |
self.assertEqual(osByte, b'\xff') # OS "unknown" (OS-independent) |
| 387 |
|
| 388 |
# Since the FNAME flag is set, the zero-terminated filename follows. |
| 389 |
# RFC 1952 specifies that this is the name of the input file, if any. |
| 390 |
# However, the gzip module defaults to storing the name of the output |
| 391 |
# file in this field. |
| 392 |
nameBytes = fRead.read(len(expectedname)) |
| 393 |
self.assertEqual(nameBytes, expectedname) |
| 394 |
|
| 395 |
# Since no other flags were set, the header ends here. |
| 396 |
# Rather than process the compressed data, let's seek to the trailer. |
| 397 |
fRead.seek(os.stat(self.filename).st_size - 8) |
| 398 |
|
| 399 |
crc32Bytes = fRead.read(4) # CRC32 of uncompressed data [data1] |
| 400 |
self.assertEqual(crc32Bytes, b'\xaf\xd7d\x83') |
| 401 |
|
| 402 |
isizeBytes = fRead.read(4) |
| 403 |
self.assertEqual(isizeBytes, struct.pack('<i', len(data1))) |
| 404 |
|
| 405 |
def test_metadata_ascii_name(self): |
| 406 |
self.filename = os_helper.TESTFN_ASCII |
| 407 |
self.test_metadata() |
| 408 |
|
| 409 |
def test_compresslevel_metadata(self): |
| 410 |
# see RFC 1952: http://www.faqs.org/rfcs/rfc1952.html |
| 411 |
# specifically, discussion of XFL in section 2.3.1 |
| 412 |
cases = [ |
| 413 |
('fast', 1, b'\x04'), |
| 414 |
('best', 9, b'\x02'), |
| 415 |
('tradeoff', 6, b'\x00'), |
| 416 |
] |
| 417 |
xflOffset = 8 |
| 418 |
|
| 419 |
for (name, level, expectedXflByte) in cases: |
| 420 |
with self.subTest(name): |
| 421 |
fWrite = gzip.GzipFile(self.filename, 'w', compresslevel=level) |
| 422 |
with fWrite: |
| 423 |
fWrite.write(data1) |
| 424 |
with open(self.filename, 'rb') as fRead: |
| 425 |
fRead.seek(xflOffset) |
| 426 |
xflByte = fRead.read(1) |
| 427 |
self.assertEqual(xflByte, expectedXflByte) |
| 428 |
|
| 429 |
def test_with_open(self): |
| 430 |
# GzipFile supports the context management protocol |
| 431 |
with gzip.GzipFile(self.filename, "wb") as f: |
| 432 |
f.write(b"xxx") |
| 433 |
f = gzip.GzipFile(self.filename, "rb") |
| 434 |
f.close() |
| 435 |
try: |
| 436 |
with f: |
| 437 |
pass |
| 438 |
except ValueError: |
| 439 |
pass |
| 440 |
else: |
| 441 |
self.fail("__enter__ on a closed file didn't raise an exception") |
| 442 |
try: |
| 443 |
with gzip.GzipFile(self.filename, "wb") as f: |
| 444 |
1/0 |
| 445 |
except ZeroDivisionError: |
| 446 |
pass |
| 447 |
else: |
| 448 |
self.fail("1/0 didn't raise an exception") |
| 449 |
|
| 450 |
def test_zero_padded_file(self): |
| 451 |
with gzip.GzipFile(self.filename, "wb") as f: |
| 452 |
f.write(data1 * 50) |
| 453 |
|
| 454 |
# Pad the file with zeroes |
| 455 |
with open(self.filename, "ab") as f: |
| 456 |
f.write(b"\x00" * 50) |
| 457 |
|
| 458 |
with gzip.GzipFile(self.filename, "rb") as f: |
| 459 |
d = f.read() |
| 460 |
self.assertEqual(d, data1 * 50, "Incorrect data in file") |
| 461 |
|
| 462 |
def test_gzip_BadGzipFile_exception(self): |
| 463 |
self.assertIsSubclass(gzip.BadGzipFile, OSError) |
| 464 |
|
| 465 |
def test_bad_gzip_file(self): |
| 466 |
with open(self.filename, 'wb') as file: |
| 467 |
file.write(data1 * 50) |
| 468 |
with gzip.GzipFile(self.filename, 'r') as file: |
| 469 |
self.assertRaises(gzip.BadGzipFile, file.readlines) |
| 470 |
|
| 471 |
def test_non_seekable_file(self): |
| 472 |
uncompressed = data1 * 50 |
| 473 |
buf = UnseekableIO() |
| 474 |
with gzip.GzipFile(fileobj=buf, mode="wb") as f: |
| 475 |
f.write(uncompressed) |
| 476 |
compressed = buf.getvalue() |
| 477 |
buf = UnseekableIO(compressed) |
| 478 |
with gzip.GzipFile(fileobj=buf, mode="rb") as f: |
| 479 |
self.assertEqual(f.read(), uncompressed) |
| 480 |
|
| 481 |
def test_peek(self): |
| 482 |
uncompressed = data1 * 200 |
| 483 |
with gzip.GzipFile(self.filename, "wb") as f: |
| 484 |
f.write(uncompressed) |
| 485 |
|
| 486 |
def sizes(): |
| 487 |
while True: |
| 488 |
for n in range(5, 50, 10): |
| 489 |
yield n |
| 490 |
|
| 491 |
with gzip.GzipFile(self.filename, "rb") as f: |
| 492 |
f.max_read_chunk = 33 |
| 493 |
nread = 0 |
| 494 |
for n in sizes(): |
| 495 |
s = f.peek(n) |
| 496 |
if s == b'': |
| 497 |
break |
| 498 |
self.assertEqual(f.read(len(s)), s) |
| 499 |
nread += len(s) |
| 500 |
self.assertEqual(f.read(100), b'') |
| 501 |
self.assertEqual(nread, len(uncompressed)) |
| 502 |
|
| 503 |
def test_textio_readlines(self): |
| 504 |
# Issue #10791: TextIOWrapper.readlines() fails when wrapping GzipFile. |
| 505 |
lines = (data1 * 50).decode("ascii").splitlines(keepends=True) |
| 506 |
self.test_write() |
| 507 |
with gzip.GzipFile(self.filename, 'r') as f: |
| 508 |
with io.TextIOWrapper(f, encoding="ascii") as t: |
| 509 |
self.assertEqual(t.readlines(), lines) |
| 510 |
|
| 511 |
def test_fileobj_with_name(self): |
| 512 |
with open(self.filename, "xb") as raw: |
| 513 |
with gzip.GzipFile(fileobj=raw, mode="x") as f: |
| 514 |
f.write(b'one') |
| 515 |
self.assertEqual(f.name, raw.name) |
| 516 |
self.assertEqual(f.fileno(), raw.fileno()) |
| 517 |
self.assertEqual(f.mode, gzip.WRITE) |
| 518 |
self.assertIs(f.readable(), False) |
| 519 |
self.assertIs(f.writable(), True) |
| 520 |
self.assertIs(f.seekable(), True) |
| 521 |
self.assertIs(f.closed, False) |
| 522 |
self.assertIs(f.closed, True) |
| 523 |
self.assertEqual(f.name, raw.name) |
| 524 |
self.assertRaises(AttributeError, f.fileno) |
| 525 |
self.assertEqual(f.mode, gzip.WRITE) |
| 526 |
self.assertIs(f.readable(), False) |
| 527 |
self.assertIs(f.writable(), True) |
| 528 |
self.assertIs(f.seekable(), True) |
| 529 |
|
| 530 |
with open(self.filename, "wb") as raw: |
| 531 |
with gzip.GzipFile(fileobj=raw, mode="w") as f: |
| 532 |
f.write(b'two') |
| 533 |
self.assertEqual(f.name, raw.name) |
| 534 |
self.assertEqual(f.fileno(), raw.fileno()) |
| 535 |
self.assertEqual(f.mode, gzip.WRITE) |
| 536 |
self.assertIs(f.readable(), False) |
| 537 |
self.assertIs(f.writable(), True) |
| 538 |
self.assertIs(f.seekable(), True) |
| 539 |
self.assertIs(f.closed, False) |
| 540 |
self.assertIs(f.closed, True) |
| 541 |
self.assertEqual(f.name, raw.name) |
| 542 |
self.assertRaises(AttributeError, f.fileno) |
| 543 |
self.assertEqual(f.mode, gzip.WRITE) |
| 544 |
self.assertIs(f.readable(), False) |
| 545 |
self.assertIs(f.writable(), True) |
| 546 |
self.assertIs(f.seekable(), True) |
| 547 |
|
| 548 |
with open(self.filename, "ab") as raw: |
| 549 |
with gzip.GzipFile(fileobj=raw, mode="a") as f: |
| 550 |
f.write(b'three') |
| 551 |
self.assertEqual(f.name, raw.name) |
| 552 |
self.assertEqual(f.fileno(), raw.fileno()) |
| 553 |
self.assertEqual(f.mode, gzip.WRITE) |
| 554 |
self.assertIs(f.readable(), False) |
| 555 |
self.assertIs(f.writable(), True) |
| 556 |
self.assertIs(f.seekable(), True) |
| 557 |
self.assertIs(f.closed, False) |
| 558 |
self.assertIs(f.closed, True) |
| 559 |
self.assertEqual(f.name, raw.name) |
| 560 |
self.assertRaises(AttributeError, f.fileno) |
| 561 |
self.assertEqual(f.mode, gzip.WRITE) |
| 562 |
self.assertIs(f.readable(), False) |
| 563 |
self.assertIs(f.writable(), True) |
| 564 |
self.assertIs(f.seekable(), True) |
| 565 |
|
| 566 |
with open(self.filename, "rb") as raw: |
| 567 |
with gzip.GzipFile(fileobj=raw, mode="r") as f: |
| 568 |
self.assertEqual(f.read(), b'twothree') |
| 569 |
self.assertEqual(f.name, raw.name) |
| 570 |
self.assertEqual(f.fileno(), raw.fileno()) |
| 571 |
self.assertEqual(f.mode, gzip.READ) |
| 572 |
self.assertIs(f.readable(), True) |
| 573 |
self.assertIs(f.writable(), False) |
| 574 |
self.assertIs(f.seekable(), True) |
| 575 |
self.assertIs(f.closed, False) |
| 576 |
self.assertIs(f.closed, True) |
| 577 |
self.assertEqual(f.name, raw.name) |
| 578 |
self.assertRaises(AttributeError, f.fileno) |
| 579 |
self.assertEqual(f.mode, gzip.READ) |
| 580 |
self.assertIs(f.readable(), True) |
| 581 |
self.assertIs(f.writable(), False) |
| 582 |
self.assertIs(f.seekable(), True) |
| 583 |
|
| 584 |
def test_fileobj_from_fdopen(self): |
| 585 |
# Issue #13781: Opening a GzipFile for writing fails when using a |
| 586 |
# fileobj created with os.fdopen(). |
| 587 |
fd = os.open(self.filename, os.O_WRONLY | os.O_CREAT | os.O_EXCL) |
| 588 |
with os.fdopen(fd, "xb") as raw: |
| 589 |
with gzip.GzipFile(fileobj=raw, mode="x") as f: |
| 590 |
f.write(b'one') |
| 591 |
self.assertEqual(f.name, '') |
| 592 |
self.assertEqual(f.fileno(), raw.fileno()) |
| 593 |
self.assertIs(f.closed, True) |
| 594 |
self.assertEqual(f.name, '') |
| 595 |
self.assertRaises(AttributeError, f.fileno) |
| 596 |
|
| 597 |
fd = os.open(self.filename, os.O_WRONLY | os.O_CREAT | os.O_TRUNC) |
| 598 |
with os.fdopen(fd, "wb") as raw: |
| 599 |
with gzip.GzipFile(fileobj=raw, mode="w") as f: |
| 600 |
f.write(b'two') |
| 601 |
self.assertEqual(f.name, '') |
| 602 |
self.assertEqual(f.fileno(), raw.fileno()) |
| 603 |
self.assertEqual(f.name, '') |
| 604 |
self.assertRaises(AttributeError, f.fileno) |
| 605 |
|
| 606 |
fd = os.open(self.filename, os.O_WRONLY | os.O_CREAT | os.O_APPEND) |
| 607 |
with os.fdopen(fd, "ab") as raw: |
| 608 |
with gzip.GzipFile(fileobj=raw, mode="a") as f: |
| 609 |
f.write(b'three') |
| 610 |
self.assertEqual(f.name, '') |
| 611 |
self.assertEqual(f.fileno(), raw.fileno()) |
| 612 |
self.assertEqual(f.name, '') |
| 613 |
self.assertRaises(AttributeError, f.fileno) |
| 614 |
|
| 615 |
fd = os.open(self.filename, os.O_RDONLY) |
| 616 |
with os.fdopen(fd, "rb") as raw: |
| 617 |
with gzip.GzipFile(fileobj=raw, mode="r") as f: |
| 618 |
self.assertEqual(f.read(), b'twothree') |
| 619 |
self.assertEqual(f.name, '') |
| 620 |
self.assertEqual(f.fileno(), raw.fileno()) |
| 621 |
self.assertEqual(f.name, '') |
| 622 |
self.assertRaises(AttributeError, f.fileno) |
| 623 |
|
| 624 |
def test_fileobj_mode(self): |
| 625 |
self.assertEqual(gzip.READ, 'rb') |
| 626 |
self.assertEqual(gzip.WRITE, 'wb') |
| 627 |
gzip.GzipFile(self.filename, "wb").close() |
| 628 |
with open(self.filename, "r+b") as f: |
| 629 |
with gzip.GzipFile(fileobj=f, mode='r') as g: |
| 630 |
self.assertEqual(g.mode, gzip.READ) |
| 631 |
with gzip.GzipFile(fileobj=f, mode='w') as g: |
| 632 |
self.assertEqual(g.mode, gzip.WRITE) |
| 633 |
with gzip.GzipFile(fileobj=f, mode='a') as g: |
| 634 |
self.assertEqual(g.mode, gzip.WRITE) |
| 635 |
with gzip.GzipFile(fileobj=f, mode='x') as g: |
| 636 |
self.assertEqual(g.mode, gzip.WRITE) |
| 637 |
with self.assertRaises(ValueError): |
| 638 |
gzip.GzipFile(fileobj=f, mode='z') |
| 639 |
for mode in "rb", "r+b": |
| 640 |
with open(self.filename, mode) as f: |
| 641 |
with gzip.GzipFile(fileobj=f) as g: |
| 642 |
self.assertEqual(g.mode, gzip.READ) |
| 643 |
for mode in "wb", "ab", "xb": |
| 644 |
if "x" in mode: |
| 645 |
os_helper.unlink(self.filename) |
| 646 |
with open(self.filename, mode) as f: |
| 647 |
with self.assertWarns(FutureWarning): |
| 648 |
g = gzip.GzipFile(fileobj=f) |
| 649 |
with g: |
| 650 |
self.assertEqual(g.mode, gzip.WRITE) |
| 651 |
|
| 652 |
def test_bytes_filename(self): |
| 653 |
str_filename = self.filename |
| 654 |
bytes_filename = os.fsencode(str_filename) |
| 655 |
with gzip.GzipFile(bytes_filename, "wb") as f: |
| 656 |
f.write(data1 * 50) |
| 657 |
self.assertEqual(f.name, bytes_filename) |
| 658 |
with gzip.GzipFile(bytes_filename, "rb") as f: |
| 659 |
self.assertEqual(f.read(), data1 * 50) |
| 660 |
self.assertEqual(f.name, bytes_filename) |
| 661 |
# Sanity check that we are actually operating on the right file. |
| 662 |
with gzip.GzipFile(str_filename, "rb") as f: |
| 663 |
self.assertEqual(f.read(), data1 * 50) |
| 664 |
self.assertEqual(f.name, str_filename) |
| 665 |
|
| 666 |
def test_fileobj_without_name(self): |
| 667 |
bio = io.BytesIO() |
| 668 |
with gzip.GzipFile(fileobj=bio, mode='wb') as f: |
| 669 |
f.write(data1 * 50) |
| 670 |
self.assertEqual(f.name, '') |
| 671 |
self.assertRaises(io.UnsupportedOperation, f.fileno) |
| 672 |
self.assertEqual(f.mode, gzip.WRITE) |
| 673 |
self.assertIs(f.readable(), False) |
| 674 |
self.assertIs(f.writable(), True) |
| 675 |
self.assertIs(f.seekable(), True) |
| 676 |
self.assertIs(f.closed, False) |
| 677 |
self.assertIs(f.closed, True) |
| 678 |
self.assertEqual(f.name, '') |
| 679 |
self.assertRaises(AttributeError, f.fileno) |
| 680 |
self.assertEqual(f.mode, gzip.WRITE) |
| 681 |
self.assertIs(f.readable(), False) |
| 682 |
self.assertIs(f.writable(), True) |
| 683 |
self.assertIs(f.seekable(), True) |
| 684 |
|
| 685 |
bio.seek(0) |
| 686 |
with gzip.GzipFile(fileobj=bio, mode='rb') as f: |
| 687 |
self.assertEqual(f.read(), data1 * 50) |
| 688 |
self.assertEqual(f.name, '') |
| 689 |
self.assertRaises(io.UnsupportedOperation, f.fileno) |
| 690 |
self.assertEqual(f.mode, gzip.READ) |
| 691 |
self.assertIs(f.readable(), True) |
| 692 |
self.assertIs(f.writable(), False) |
| 693 |
self.assertIs(f.seekable(), True) |
| 694 |
self.assertIs(f.closed, False) |
| 695 |
self.assertIs(f.closed, True) |
| 696 |
self.assertEqual(f.name, '') |
| 697 |
self.assertRaises(AttributeError, f.fileno) |
| 698 |
self.assertEqual(f.mode, gzip.READ) |
| 699 |
self.assertIs(f.readable(), True) |
| 700 |
self.assertIs(f.writable(), False) |
| 701 |
self.assertIs(f.seekable(), True) |
| 702 |
|
| 703 |
def test_fileobj_and_filename(self): |
| 704 |
filename2 = self.filename + 'new' |
| 705 |
with (open(self.filename, 'wb') as fileobj, |
| 706 |
gzip.GzipFile(fileobj=fileobj, filename=filename2, mode='wb') as f): |
| 707 |
f.write(data1 * 50) |
| 708 |
self.assertEqual(f.name, filename2) |
| 709 |
with (open(self.filename, 'rb') as fileobj, |
| 710 |
gzip.GzipFile(fileobj=fileobj, filename=filename2, mode='rb') as f): |
| 711 |
self.assertEqual(f.read(), data1 * 50) |
| 712 |
self.assertEqual(f.name, filename2) |
| 713 |
# Sanity check that we are actually operating on the right file. |
| 714 |
with gzip.GzipFile(self.filename, 'rb') as f: |
| 715 |
self.assertEqual(f.read(), data1 * 50) |
| 716 |
self.assertEqual(f.name, self.filename) |
| 717 |
|
| 718 |
def test_decompress_limited(self): |
| 719 |
"""Decompressed data buffering should be limited""" |
| 720 |
bomb = gzip.compress(b'\0' * int(2e6), compresslevel=9) |
| 721 |
self.assertLess(len(bomb), io.DEFAULT_BUFFER_SIZE) |
| 722 |
|
| 723 |
bomb = io.BytesIO(bomb) |
| 724 |
decomp = gzip.GzipFile(fileobj=bomb) |
| 725 |
self.assertEqual(decomp.read(1), b'\0') |
| 726 |
max_decomp = 1 + io.DEFAULT_BUFFER_SIZE |
| 727 |
self.assertLessEqual(decomp._buffer.raw.tell(), max_decomp, |
| 728 |
"Excessive amount of data was decompressed") |
| 729 |
|
| 730 |
# Testing compress/decompress shortcut functions |
| 731 |
|
| 732 |
def test_compress(self): |
| 733 |
for data in [data1, data2]: |
| 734 |
for args in [(), (1,), (6,), (9,)]: |
| 735 |
datac = gzip.compress(data, *args) |
| 736 |
self.assertEqual(type(datac), bytes) |
| 737 |
with gzip.GzipFile(fileobj=io.BytesIO(datac), mode="rb") as f: |
| 738 |
self.assertEqual(f.read(), data) |
| 739 |
|
| 740 |
def test_compress_mtime(self): |
| 741 |
mtime = 123456789 |
| 742 |
for data in [data1, data2]: |
| 743 |
for args in [(), (1,), (6,), (9,)]: |
| 744 |
with self.subTest(data=data, args=args): |
| 745 |
datac = gzip.compress(data, *args, mtime=mtime) |
| 746 |
self.assertEqual(type(datac), bytes) |
| 747 |
with gzip.GzipFile(fileobj=io.BytesIO(datac), mode="rb") as f: |
| 748 |
f.read(1) # to set mtime attribute |
| 749 |
self.assertEqual(f.mtime, mtime) |
| 750 |
|
| 751 |
def test_compress_mtime_default(self): |
| 752 |
# test for gh-125260 |
| 753 |
datac = gzip.compress(data1, mtime=0) |
| 754 |
datac2 = gzip.compress(data1) |
| 755 |
self.assertEqual(datac, datac2) |
| 756 |
datac3 = gzip.compress(data1, mtime=None) |
| 757 |
self.assertNotEqual(datac, datac3) |
| 758 |
with gzip.GzipFile(fileobj=io.BytesIO(datac3), mode="rb") as f: |
| 759 |
f.read(1) # to set mtime attribute |
| 760 |
self.assertGreater(f.mtime, 1) |
| 761 |
|
| 762 |
def test_compress_correct_level(self): |
| 763 |
for mtime in (0, 42): |
| 764 |
with self.subTest(mtime=mtime): |
| 765 |
nocompress = gzip.compress(data1, compresslevel=0, mtime=mtime) |
| 766 |
yescompress = gzip.compress(data1, compresslevel=1, mtime=mtime) |
| 767 |
self.assertIn(data1, nocompress) |
| 768 |
self.assertNotIn(data1, yescompress) |
| 769 |
|
| 770 |
def test_issue112346(self): |
| 771 |
# The OS byte should be 255, this should not change between Python versions. |
| 772 |
for mtime in (0, 42): |
| 773 |
with self.subTest(mtime=mtime): |
| 774 |
compress = gzip.compress(data1, compresslevel=1, mtime=mtime) |
| 775 |
self.assertEqual( |
| 776 |
struct.unpack("<IxB", compress[4:10]), |
| 777 |
(mtime, 255), |
| 778 |
"Gzip header does not properly set either mtime or OS byte." |
| 779 |
) |
| 780 |
|
| 781 |
def test_decompress(self): |
| 782 |
for data in (data1, data2): |
| 783 |
buf = io.BytesIO() |
| 784 |
with gzip.GzipFile(fileobj=buf, mode="wb") as f: |
| 785 |
f.write(data) |
| 786 |
self.assertEqual(gzip.decompress(buf.getvalue()), data) |
| 787 |
# Roundtrip with compress |
| 788 |
datac = gzip.compress(data) |
| 789 |
self.assertEqual(gzip.decompress(datac), data) |
| 790 |
|
| 791 |
def test_decompress_truncated_trailer(self): |
| 792 |
compressed_data = gzip.compress(data1) |
| 793 |
self.assertRaises(EOFError, gzip.decompress, compressed_data[:-4]) |
| 794 |
|
| 795 |
def test_decompress_missing_trailer(self): |
| 796 |
compressed_data = gzip.compress(data1) |
| 797 |
self.assertRaises(EOFError, gzip.decompress, compressed_data[:-8]) |
| 798 |
|
| 799 |
def test_read_truncated(self): |
| 800 |
data = data1*50 |
| 801 |
# Drop the CRC (4 bytes) and file size (4 bytes). |
| 802 |
truncated = gzip.compress(data)[:-8] |
| 803 |
with gzip.GzipFile(fileobj=io.BytesIO(truncated)) as f: |
| 804 |
self.assertRaises(EOFError, f.read) |
| 805 |
with gzip.GzipFile(fileobj=io.BytesIO(truncated)) as f: |
| 806 |
self.assertEqual(f.read(len(data)), data) |
| 807 |
self.assertRaises(EOFError, f.read, 1) |
| 808 |
# Incomplete 10-byte header. |
| 809 |
for i in range(2, 10): |
| 810 |
with gzip.GzipFile(fileobj=io.BytesIO(truncated[:i])) as f: |
| 811 |
self.assertRaises(EOFError, f.read, 1) |
| 812 |
|
| 813 |
def test_read_with_extra(self): |
| 814 |
# Gzip data with an extra field |
| 815 |
gzdata = (b'\x1f\x8b\x08\x04\xb2\x17cQ\x02\xff' |
| 816 |
b'\x05\x00Extra' |
| 817 |
b'\x0bI-.\x01\x002\xd1Mx\x04\x00\x00\x00') |
| 818 |
with gzip.GzipFile(fileobj=io.BytesIO(gzdata)) as f: |
| 819 |
self.assertEqual(f.read(), b'Test') |
| 820 |
|
| 821 |
def test_prepend_error(self): |
| 822 |
# See issue #20875 |
| 823 |
with gzip.open(self.filename, "wb") as f: |
| 824 |
f.write(data1) |
| 825 |
with gzip.open(self.filename, "rb") as f: |
| 826 |
f._buffer.raw._fp.prepend() |
| 827 |
|
| 828 |
def test_issue44439(self): |
| 829 |
q = array.array('Q', [1, 2, 3, 4, 5]) |
| 830 |
LENGTH = len(q) * q.itemsize |
| 831 |
|
| 832 |
with gzip.GzipFile(fileobj=io.BytesIO(), mode='w') as f: |
| 833 |
self.assertEqual(f.write(q), LENGTH) |
| 834 |
self.assertEqual(f.tell(), LENGTH) |
| 835 |
|
| 836 |
def test_flush_flushes_compressor(self): |
| 837 |
# See issue GH-105808. |
| 838 |
b = io.BytesIO() |
| 839 |
message = b"important message here." |
| 840 |
with gzip.GzipFile(fileobj=b, mode='w') as f: |
| 841 |
f.write(message) |
| 842 |
f.flush() |
| 843 |
partial_data = b.getvalue() |
| 844 |
full_data = b.getvalue() |
| 845 |
self.assertEqual(gzip.decompress(full_data), message) |
| 846 |
# The partial data should contain the gzip header and the complete |
| 847 |
# message, but not the end-of-stream markers (so we can't just |
| 848 |
# decompress it directly). |
| 849 |
with self.assertRaises(EOFError): |
| 850 |
gzip.decompress(partial_data) |
| 851 |
d = zlib.decompressobj(wbits=-zlib.MAX_WBITS) |
| 852 |
f = io.BytesIO(partial_data) |
| 853 |
gzip._read_gzip_header(f) |
| 854 |
read_message = d.decompress(f.read()) |
| 855 |
self.assertEqual(read_message, message) |
| 856 |
|
| 857 |
def test_flush_modes(self): |
| 858 |
# Make sure the argument to flush is properly passed to the |
| 859 |
# zlib.compressobj; see issue GH-105808. |
| 860 |
class FakeCompressor: |
| 861 |
def __init__(self): |
| 862 |
self.modes = [] |
| 863 |
def compress(self, data): |
| 864 |
return b'' |
| 865 |
def flush(self, mode=-1): |
| 866 |
self.modes.append(mode) |
| 867 |
return b'' |
| 868 |
b = io.BytesIO() |
| 869 |
fc = FakeCompressor() |
| 870 |
with gzip.GzipFile(fileobj=b, mode='w') as f: |
| 871 |
f.compress = fc |
| 872 |
f.flush() |
| 873 |
f.flush(50) |
| 874 |
f.flush(zlib_mode=100) |
| 875 |
# The implicit close will also flush the compressor. |
| 876 |
expected_modes = [ |
| 877 |
zlib.Z_SYNC_FLUSH, |
| 878 |
50, |
| 879 |
100, |
| 880 |
-1, |
| 881 |
] |
| 882 |
self.assertEqual(fc.modes, expected_modes) |
| 883 |
|
| 884 |
def test_write_seek_write(self): |
| 885 |
# Make sure that offset is up-to-date before seeking |
| 886 |
# See issue GH-108111 |
| 887 |
b = io.BytesIO() |
| 888 |
message = b"important message here." |
| 889 |
with gzip.GzipFile(fileobj=b, mode='w') as f: |
| 890 |
f.write(message) |
| 891 |
f.seek(len(message)) |
| 892 |
f.write(message) |
| 893 |
data = b.getvalue() |
| 894 |
self.assertEqual(gzip.decompress(data), message * 2) |
| 895 |
|
| 896 |
|
| 897 |
def test_refloop_unraisable(self): |
| 898 |
# Ensure a GzipFile referring to a temporary fileobj deletes cleanly. |
| 899 |
# Previously an unraisable exception would occur on close because the |
| 900 |
# fileobj would be closed before the GzipFile as the result of a |
| 901 |
# reference loop. See issue gh-129726 |
| 902 |
with catch_unraisable_exception() as cm: |
| 903 |
with self.assertWarns(ResourceWarning): |
| 904 |
gzip.GzipFile(fileobj=io.BytesIO(), mode="w") |
| 905 |
gc.collect() |
| 906 |
self.assertIsNone(cm.unraisable) |
| 907 |
|
| 908 |
|
| 909 |
class TestOpen(BaseTest): |
| 910 |
def test_binary_modes(self): |
| 911 |
uncompressed = data1 * 50 |
| 912 |
|
| 913 |
with gzip.open(self.filename, "wb") as f: |
| 914 |
f.write(uncompressed) |
| 915 |
with open(self.filename, "rb") as f: |
| 916 |
file_data = gzip.decompress(f.read()) |
| 917 |
self.assertEqual(file_data, uncompressed) |
| 918 |
|
| 919 |
with gzip.open(self.filename, "rb") as f: |
| 920 |
self.assertEqual(f.read(), uncompressed) |
| 921 |
|
| 922 |
with gzip.open(self.filename, "ab") as f: |
| 923 |
f.write(uncompressed) |
| 924 |
with open(self.filename, "rb") as f: |
| 925 |
file_data = gzip.decompress(f.read()) |
| 926 |
self.assertEqual(file_data, uncompressed * 2) |
| 927 |
|
| 928 |
with self.assertRaises(FileExistsError): |
| 929 |
gzip.open(self.filename, "xb") |
| 930 |
os_helper.unlink(self.filename) |
| 931 |
with gzip.open(self.filename, "xb") as f: |
| 932 |
f.write(uncompressed) |
| 933 |
with open(self.filename, "rb") as f: |
| 934 |
file_data = gzip.decompress(f.read()) |
| 935 |
self.assertEqual(file_data, uncompressed) |
| 936 |
|
| 937 |
def test_pathlike_file(self): |
| 938 |
filename = os_helper.FakePath(self.filename) |
| 939 |
with gzip.open(filename, "wb") as f: |
| 940 |
f.write(data1 * 50) |
| 941 |
self.assertEqual(f.name, self.filename) |
| 942 |
with gzip.open(filename, "ab") as f: |
| 943 |
f.write(data1) |
| 944 |
self.assertEqual(f.name, self.filename) |
| 945 |
with gzip.open(filename) as f: |
| 946 |
self.assertEqual(f.read(), data1 * 51) |
| 947 |
self.assertEqual(f.name, self.filename) |
| 948 |
|
| 949 |
def test_implicit_binary_modes(self): |
| 950 |
# Test implicit binary modes (no "b" or "t" in mode string). |
| 951 |
uncompressed = data1 * 50 |
| 952 |
|
| 953 |
with gzip.open(self.filename, "w") as f: |
| 954 |
f.write(uncompressed) |
| 955 |
with open(self.filename, "rb") as f: |
| 956 |
file_data = gzip.decompress(f.read()) |
| 957 |
self.assertEqual(file_data, uncompressed) |
| 958 |
|
| 959 |
with gzip.open(self.filename, "r") as f: |
| 960 |
self.assertEqual(f.read(), uncompressed) |
| 961 |
|
| 962 |
with gzip.open(self.filename, "a") as f: |
| 963 |
f.write(uncompressed) |
| 964 |
with open(self.filename, "rb") as f: |
| 965 |
file_data = gzip.decompress(f.read()) |
| 966 |
self.assertEqual(file_data, uncompressed * 2) |
| 967 |
|
| 968 |
with self.assertRaises(FileExistsError): |
| 969 |
gzip.open(self.filename, "x") |
| 970 |
os_helper.unlink(self.filename) |
| 971 |
with gzip.open(self.filename, "x") as f: |
| 972 |
f.write(uncompressed) |
| 973 |
with open(self.filename, "rb") as f: |
| 974 |
file_data = gzip.decompress(f.read()) |
| 975 |
self.assertEqual(file_data, uncompressed) |
| 976 |
|
| 977 |
def test_text_modes(self): |
| 978 |
uncompressed = data1.decode("ascii") * 50 |
| 979 |
uncompressed_raw = uncompressed.replace("\n", os.linesep) |
| 980 |
with gzip.open(self.filename, "wt", encoding="ascii") as f: |
| 981 |
f.write(uncompressed) |
| 982 |
with open(self.filename, "rb") as f: |
| 983 |
file_data = gzip.decompress(f.read()).decode("ascii") |
| 984 |
self.assertEqual(file_data, uncompressed_raw) |
| 985 |
with gzip.open(self.filename, "rt", encoding="ascii") as f: |
| 986 |
self.assertEqual(f.read(), uncompressed) |
| 987 |
with gzip.open(self.filename, "at", encoding="ascii") as f: |
| 988 |
f.write(uncompressed) |
| 989 |
with open(self.filename, "rb") as f: |
| 990 |
file_data = gzip.decompress(f.read()).decode("ascii") |
| 991 |
self.assertEqual(file_data, uncompressed_raw * 2) |
| 992 |
|
| 993 |
def test_fileobj(self): |
| 994 |
uncompressed_bytes = data1 * 50 |
| 995 |
uncompressed_str = uncompressed_bytes.decode("ascii") |
| 996 |
compressed = gzip.compress(uncompressed_bytes) |
| 997 |
with gzip.open(io.BytesIO(compressed), "r") as f: |
| 998 |
self.assertEqual(f.read(), uncompressed_bytes) |
| 999 |
with gzip.open(io.BytesIO(compressed), "rb") as f: |
| 1000 |
self.assertEqual(f.read(), uncompressed_bytes) |
| 1001 |
with gzip.open(io.BytesIO(compressed), "rt", encoding="ascii") as f: |
| 1002 |
self.assertEqual(f.read(), uncompressed_str) |
| 1003 |
|
| 1004 |
def test_bad_params(self): |
| 1005 |
# Test invalid parameter combinations. |
| 1006 |
with self.assertRaises(TypeError): |
| 1007 |
gzip.open(123.456) |
| 1008 |
with self.assertRaises(ValueError): |
| 1009 |
gzip.open(self.filename, "wbt") |
| 1010 |
with self.assertRaises(ValueError): |
| 1011 |
gzip.open(self.filename, "xbt") |
| 1012 |
with self.assertRaises(ValueError): |
| 1013 |
gzip.open(self.filename, "rb", encoding="utf-8") |
| 1014 |
with self.assertRaises(ValueError): |
| 1015 |
gzip.open(self.filename, "rb", errors="ignore") |
| 1016 |
with self.assertRaises(ValueError): |
| 1017 |
gzip.open(self.filename, "rb", newline="\n") |
| 1018 |
|
| 1019 |
def test_encoding(self): |
| 1020 |
# Test non-default encoding. |
| 1021 |
uncompressed = data1.decode("ascii") * 50 |
| 1022 |
uncompressed_raw = uncompressed.replace("\n", os.linesep) |
| 1023 |
with gzip.open(self.filename, "wt", encoding="utf-16") as f: |
| 1024 |
f.write(uncompressed) |
| 1025 |
with open(self.filename, "rb") as f: |
| 1026 |
file_data = gzip.decompress(f.read()).decode("utf-16") |
| 1027 |
self.assertEqual(file_data, uncompressed_raw) |
| 1028 |
with gzip.open(self.filename, "rt", encoding="utf-16") as f: |
| 1029 |
self.assertEqual(f.read(), uncompressed) |
| 1030 |
|
| 1031 |
def test_encoding_error_handler(self): |
| 1032 |
# Test with non-default encoding error handler. |
| 1033 |
with gzip.open(self.filename, "wb") as f: |
| 1034 |
f.write(b"foo\xffbar") |
| 1035 |
with gzip.open(self.filename, "rt", encoding="ascii", errors="ignore") \ |
| 1036 |
as f: |
| 1037 |
self.assertEqual(f.read(), "foobar") |
| 1038 |
|
| 1039 |
def test_newline(self): |
| 1040 |
# Test with explicit newline (universal newline mode disabled). |
| 1041 |
uncompressed = data1.decode("ascii") * 50 |
| 1042 |
with gzip.open(self.filename, "wt", encoding="ascii", newline="\n") as f: |
| 1043 |
f.write(uncompressed) |
| 1044 |
with gzip.open(self.filename, "rt", encoding="ascii", newline="\r") as f: |
| 1045 |
self.assertEqual(f.readlines(), [uncompressed]) |
| 1046 |
|
| 1047 |
|
| 1048 |
def create_and_remove_directory(directory): |
| 1049 |
def decorator(function): |
| 1050 |
@functools.wraps(function) |
| 1051 |
def wrapper(*args, **kwargs): |
| 1052 |
os.makedirs(directory) |
| 1053 |
try: |
| 1054 |
return function(*args, **kwargs) |
| 1055 |
finally: |
| 1056 |
os_helper.rmtree(directory) |
| 1057 |
return wrapper |
| 1058 |
return decorator |
| 1059 |
|
| 1060 |
|
| 1061 |
class TestCommandLine(unittest.TestCase): |
| 1062 |
data = b'This is a simple test with gzip' |
| 1063 |
|
| 1064 |
@requires_subprocess() |
| 1065 |
def test_decompress_stdin_stdout(self): |
| 1066 |
with io.BytesIO() as bytes_io: |
| 1067 |
with gzip.GzipFile(fileobj=bytes_io, mode='wb') as gzip_file: |
| 1068 |
gzip_file.write(self.data) |
| 1069 |
|
| 1070 |
args = sys.executable, '-m', 'gzip', '-d' |
| 1071 |
with Popen(args, stdin=PIPE, stdout=PIPE, stderr=PIPE) as proc: |
| 1072 |
out, err = proc.communicate(bytes_io.getvalue()) |
| 1073 |
|
| 1074 |
self.assertEqual(err, b'') |
| 1075 |
self.assertEqual(out, self.data) |
| 1076 |
|
| 1077 |
@create_and_remove_directory(TEMPDIR) |
| 1078 |
def test_decompress_infile_outfile(self): |
| 1079 |
gzipname = os.path.join(TEMPDIR, 'testgzip.gz') |
| 1080 |
self.assertFalse(os.path.exists(gzipname)) |
| 1081 |
|
| 1082 |
with gzip.open(gzipname, mode='wb') as fp: |
| 1083 |
fp.write(self.data) |
| 1084 |
rc, out, err = assert_python_ok('-m', 'gzip', '-d', gzipname) |
| 1085 |
|
| 1086 |
with open(os.path.join(TEMPDIR, "testgzip"), "rb") as gunziped: |
| 1087 |
self.assertEqual(gunziped.read(), self.data) |
| 1088 |
|
| 1089 |
self.assertTrue(os.path.exists(gzipname)) |
| 1090 |
self.assertEqual(rc, 0) |
| 1091 |
self.assertEqual(out, b'') |
| 1092 |
self.assertEqual(err, b'') |
| 1093 |
|
| 1094 |
def test_decompress_infile_outfile_error(self): |
| 1095 |
rc, out, err = assert_python_failure('-m', 'gzip', '-d', 'thisisatest.out') |
| 1096 |
self.assertEqual(b"filename doesn't end in .gz: 'thisisatest.out'", err.strip()) |
| 1097 |
self.assertEqual(rc, 1) |
| 1098 |
self.assertEqual(out, b'') |
| 1099 |
|
| 1100 |
@requires_subprocess() |
| 1101 |
@create_and_remove_directory(TEMPDIR) |
| 1102 |
def test_compress_stdin_outfile(self): |
| 1103 |
args = sys.executable, '-m', 'gzip' |
| 1104 |
with Popen(args, stdin=PIPE, stdout=PIPE, stderr=PIPE) as proc: |
| 1105 |
out, err = proc.communicate(self.data) |
| 1106 |
|
| 1107 |
self.assertEqual(err, b'') |
| 1108 |
self.assertEqual(out[:2], b"\x1f\x8b") |
| 1109 |
|
| 1110 |
@create_and_remove_directory(TEMPDIR) |
| 1111 |
def test_compress_infile_outfile_default(self): |
| 1112 |
local_testgzip = os.path.join(TEMPDIR, 'testgzip') |
| 1113 |
gzipname = local_testgzip + '.gz' |
| 1114 |
self.assertFalse(os.path.exists(gzipname)) |
| 1115 |
|
| 1116 |
with open(local_testgzip, 'wb') as fp: |
| 1117 |
fp.write(self.data) |
| 1118 |
|
| 1119 |
rc, out, err = assert_python_ok('-m', 'gzip', local_testgzip) |
| 1120 |
|
| 1121 |
self.assertTrue(os.path.exists(gzipname)) |
| 1122 |
self.assertEqual(out, b'') |
| 1123 |
self.assertEqual(err, b'') |
| 1124 |
|
| 1125 |
@create_and_remove_directory(TEMPDIR) |
| 1126 |
def test_compress_infile_outfile(self): |
| 1127 |
for compress_level in ('--fast', '--best'): |
| 1128 |
with self.subTest(compress_level=compress_level): |
| 1129 |
local_testgzip = os.path.join(TEMPDIR, 'testgzip') |
| 1130 |
gzipname = local_testgzip + '.gz' |
| 1131 |
self.assertFalse(os.path.exists(gzipname)) |
| 1132 |
|
| 1133 |
with open(local_testgzip, 'wb') as fp: |
| 1134 |
fp.write(self.data) |
| 1135 |
|
| 1136 |
rc, out, err = assert_python_ok('-m', 'gzip', compress_level, local_testgzip) |
| 1137 |
|
| 1138 |
self.assertTrue(os.path.exists(gzipname)) |
| 1139 |
self.assertEqual(out, b'') |
| 1140 |
self.assertEqual(err, b'') |
| 1141 |
os.remove(gzipname) |
| 1142 |
self.assertFalse(os.path.exists(gzipname)) |
| 1143 |
|
| 1144 |
def test_compress_fast_best_are_exclusive(self): |
| 1145 |
rc, out, err = assert_python_failure('-m', 'gzip', '--fast', '--best') |
| 1146 |
self.assertIn(b"error: argument --best: not allowed with argument --fast", err) |
| 1147 |
self.assertEqual(out, b'') |
| 1148 |
|
| 1149 |
def test_decompress_cannot_have_flags_compression(self): |
| 1150 |
rc, out, err = assert_python_failure('-m', 'gzip', '--fast', '-d') |
| 1151 |
self.assertIn(b'error: argument -d/--decompress: not allowed with argument --fast', err) |
| 1152 |
self.assertEqual(out, b'') |
| 1153 |
|
| 1154 |
|
| 1155 |
if __name__ == "__main__": |
| 1156 |
unittest.main() |