tarname = tarname

suffix = ''

open = io.FileIO

taropen = tarfile.TarFile.taropen

@property

def mode(self):

tarname = gzipname

suffix = 'gz'

open = gzip.GzipFile if gzip else None

tarname = bz2name

suffix = 'bz2'

open = bz2.BZ2File if bz2 else None

tarname = xzname

suffix = 'xz'

open = lzma.LZMAFile if lzma else None

tarname = zstname

suffix = 'zst'

open = zstd.ZstdFile if zstd else None

prefix = "r:"

def setUp(self):

self.tar = tarfile.open(self.tarname, mode=self.mode,

encoding="iso8859-1")

def tearDown(self):

# Only needs to change how the tarfile is opened to set

# stream mode

def setUp(self):

self.tar = tarfile.open(self.tarname, mode=self.mode,

encoding="iso8859-1",

def test_fileobj_regular_file(self):

tarinfo = self.tar.getmember("ustar/regtype")

with self.tar.extractfile(tarinfo) as fobj:

data = fobj.read()

self.assertEqual(len(data), tarinfo.size,

"regular file extraction failed")

self.assertEqual(sha256sum(data), sha256_regtype,

"regular file extraction failed")

def test_fileobj_readlines(self):

self.tar.extract("ustar/regtype", TEMPDIR, filter='data')

tarinfo = self.tar.getmember("ustar/regtype")

with open(os.path.join(TEMPDIR, "ustar/regtype"), "r") as fobj1:

lines1 = fobj1.readlines()

with self.tar.extractfile(tarinfo) as fobj:

fobj2 = io.TextIOWrapper(fobj)

lines2 = fobj2.readlines()

self.assertEqual(lines1, lines2,

"fileobj.readlines() failed")

self.assertEqual(len(lines2), 114,

"fileobj.readlines() failed")

self.assertEqual(lines2[83],

"I will gladly admit that Python is not the fastest "

"running scripting language.\n",

"fileobj.readlines() failed")

def test_fileobj_iter(self):

self.tar.extract("ustar/regtype", TEMPDIR, filter='data')

tarinfo = self.tar.getmember("ustar/regtype")

with open(os.path.join(TEMPDIR, "ustar/regtype"), "r") as fobj1:

lines1 = fobj1.readlines()

with self.tar.extractfile(tarinfo) as fobj2:

lines2 = list(io.TextIOWrapper(fobj2))

self.assertEqual(lines1, lines2,

"fileobj.__iter__() failed")

def test_fileobj_seek(self):

self.tar.extract("ustar/regtype", TEMPDIR,

filter='data')

with open(os.path.join(TEMPDIR, "ustar/regtype"), "rb") as fobj:

data = fobj.read()

tarinfo = self.tar.getmember("ustar/regtype")

with self.tar.extractfile(tarinfo) as fobj:

text = fobj.read()

fobj.seek(0)

self.assertEqual(0, fobj.tell(),

"seek() to file's start failed")

fobj.seek(2048, 0)

self.assertEqual(2048, fobj.tell(),

"seek() to absolute position failed")

fobj.seek(-1024, 1)

self.assertEqual(1024, fobj.tell(),

"seek() to negative relative position failed")

fobj.seek(1024, 1)

self.assertEqual(2048, fobj.tell(),

"seek() to positive relative position failed")

s = fobj.read(10)

self.assertEqual(s, data[2048:2058],

"read() after seek failed")

fobj.seek(0, 2)

self.assertEqual(tarinfo.size, fobj.tell(),

"seek() to file's end failed")

self.assertEqual(fobj.read(), b"",

"read() at file's end did not return empty string")

fobj.seek(-tarinfo.size, 2)

self.assertEqual(0, fobj.tell(),

"relative seek() to file's end failed")

fobj.seek(512)

s1 = fobj.readlines()

fobj.seek(512)

s2 = fobj.readlines()

self.assertEqual(s1, s2,

"readlines() after seek failed")

fobj.seek(0)

self.assertEqual(len(fobj.readline()), fobj.tell(),

"tell() after readline() failed")

fobj.seek(512)

self.assertEqual(len(fobj.readline()) + 512, fobj.tell(),

"tell() after seek() and readline() failed")

fobj.seek(0)

line = fobj.readline()

self.assertEqual(fobj.read(), data[len(line):],

"read() after readline() failed")

def test_fileobj_text(self):

with self.tar.extractfile("ustar/regtype") as fobj:

fobj = io.TextIOWrapper(fobj)

data = fobj.read().encode("iso8859-1")

self.assertEqual(sha256sum(data), sha256_regtype)

try:

fobj.seek(100)

except AttributeError:

# Issue #13815: seek() complained about a missing

# flush() method.

self.fail("seeking failed in text mode")

# Test if symbolic and hard links are resolved by extractfile(). The

# test link members each point to a regular member whose data is

# supposed to be exported.

def _test_fileobj_link(self, lnktype, regtype):

with self.tar.extractfile(lnktype) as a, \

self.tar.extractfile(regtype) as b:

self.assertEqual(a.name, b.name)

def test_fileobj_link1(self):

self._test_fileobj_link("ustar/lnktype", "ustar/regtype")

def test_fileobj_link2(self):

self._test_fileobj_link("./ustar/linktest2/lnktype",

"ustar/linktest1/regtype")

def test_fileobj_symlink1(self):

self._test_fileobj_link("ustar/symtype", "ustar/regtype")

def test_fileobj_symlink2(self):

self._test_fileobj_link("./ustar/linktest2/symtype",

"ustar/linktest1/regtype")

def test_issue14160(self):

self._test_fileobj_link("symtype2", "ustar/regtype")

def test_add_dir_getmember(self):

# bpo-21987

self.add_dir_and_getmember('bar')

self.add_dir_and_getmember('a'*101)

@unittest.skipUnless(hasattr(os, "getuid") and hasattr(os, "getgid"),

def add_dir_and_getmember(self, name):

def filter(tarinfo):

tarinfo.uid = tarinfo.gid = 100

return tarinfo

with os_helper.temp_cwd():

with tarfile.open(tmpname, 'w') as tar:

tar.format = tarfile.USTAR_FORMAT

try:

os.mkdir(name)

tar.add(name, filter=filter)

finally:

os.rmdir(name)

with tarfile.open(tmpname) as tar:

self.assertEqual(

tar.getmember(name),

tar.getmember(name + '/')

# Override setUp to use default encoding (UTF-8)

def setUp(self):

self.tar = tarfile.open(self.tarname, mode=self.mode)

def test_list(self):

tio = io.TextIOWrapper(io.BytesIO(), 'ascii', newline='\n')

with support.swap_attr(sys, 'stdout', tio):

self.tar.list(verbose=False)

out = tio.detach().getvalue()

self.assertIn(b'ustar/conttype', out)

self.assertIn(b'ustar/regtype', out)

self.assertIn(b'ustar/lnktype', out)

self.assertIn(b'ustar' + (b'/12345' * 40) + b'67/longname', out)

self.assertIn(b'./ustar/linktest2/symtype', out)

self.assertIn(b'./ustar/linktest2/lnktype', out)

# Make sure it puts trailing slash for directory

self.assertIn(b'ustar/dirtype/', out)

self.assertIn(b'ustar/dirtype-with-size/', out)

# Make sure it is able to print unencodable characters

def conv(b):

s = b.decode(self.tar.encoding, 'surrogateescape')

return s.encode('ascii', 'backslashreplace')

self.assertIn(conv(b'ustar/umlauts-\xc4\xd6\xdc\xe4\xf6\xfc\xdf'), out)

self.assertIn(conv(b'misc/regtype-hpux-signed-chksum-'

b'\xc4\xd6\xdc\xe4\xf6\xfc\xdf'), out)

self.assertIn(conv(b'misc/regtype-old-v7-signed-chksum-'

b'\xc4\xd6\xdc\xe4\xf6\xfc\xdf'), out)

self.assertIn(conv(b'pax/bad-pax-\xe4\xf6\xfc'), out)

self.assertIn(conv(b'pax/hdrcharset-\xe4\xf6\xfc'), out)

# Make sure it prints files separated by one newline without any

# 'ls -l'-like accessories if verbose flag is not being used

# ...

# ustar/conttype

# ustar/regtype

# ...

self.assertRegex(out, br'ustar/conttype ?\r?\n'

br'ustar/regtype ?\r?\n')

# Make sure it does not print the source of link without verbose flag

self.assertNotIn(b'link to', out)

self.assertNotIn(b'->', out)

def test_list_verbose(self):

tio = io.TextIOWrapper(io.BytesIO(), 'ascii', newline='\n')

with support.swap_attr(sys, 'stdout', tio):

self.tar.list(verbose=True)

out = tio.detach().getvalue()

# Make sure it prints files separated by one newline with 'ls -l'-like

# accessories if verbose flag is being used

# ...

# ?rw-r--r-- tarfile/tarfile 7011 2003-01-06 07:19:43 ustar/conttype

# -rw-r--r-- tarfile/tarfile 7011 2003-01-06 07:19:43 ustar/regtype

# drwxr-xr-x tarfile/tarfile 0 2003-01-05 15:19:43 ustar/dirtype/

# ...

#

# Array of values to modify the regex below:

# ((file_type, file_permissions, file_length), ...)

type_perm_lengths = (

(br'\?', b'rw-r--r--', b'7011'), (b'-', b'rw-r--r--', b'7011'),

(b'd', b'rwxr-xr-x', b'0'), (b'd', b'rwxr-xr-x', b'255'),

(br'\?', b'rw-r--r--', b'0'), (b'l', b'rwxrwxrwx', b'0'),

(b'b', b'rw-rw----', b'3,0'), (b'c', b'rw-rw-rw-', b'1,3'),

(b'p', b'rw-r--r--', b'0'))

self.assertRegex(out, b''.join(

[(tp + (br'%s tarfile/tarfile\s+%s ' % (perm, ln) +

br'\d{4}-\d\d-\d\d\s+\d\d:\d\d:\d\d '

br'ustar/\w+type[/>\sa-z-]*\n')) for tp, perm, ln

in type_perm_lengths]))

# Make sure it prints the source of link with verbose flag

self.assertIn(b'ustar/symtype -> regtype', out)

self.assertIn(b'./ustar/linktest2/symtype -> ../linktest1/regtype', out)

self.assertIn(b'./ustar/linktest2/lnktype link to '

b'./ustar/linktest1/regtype', out)

self.assertIn(b'gnu' + (b'/123' * 125) + b'/longlink link to gnu' +

(b'/123' * 125) + b'/longname', out)

self.assertIn(b'pax' + (b'/123' * 125) + b'/longlink link to pax' +

(b'/123' * 125) + b'/longname', out)

def test_list_members(self):

tio = io.TextIOWrapper(io.BytesIO(), 'ascii', newline='\n')

def members(tar):

for tarinfo in tar.getmembers():

if 'reg' in tarinfo.name:

yield tarinfo

with support.swap_attr(sys, 'stdout', tio):

self.tar.list(verbose=False, members=members(self.tar))

out = tio.detach().getvalue()

self.assertIn(b'ustar/regtype', out)

def test_is_tarfile_erroneous(self):

with open(tmpname, "wb"):

pass

# is_tarfile works on filenames

self.assertFalse(tarfile.is_tarfile(tmpname))

# is_tarfile works on path-like objects

self.assertFalse(tarfile.is_tarfile(os_helper.FakePath(tmpname)))

# is_tarfile works on file objects

with open(tmpname, "rb") as fobj:

self.assertFalse(tarfile.is_tarfile(fobj))

# is_tarfile works on file-like objects

self.assertFalse(tarfile.is_tarfile(io.BytesIO(b"invalid")))

def test_is_tarfile_valid(self):

# is_tarfile works on filenames

self.assertTrue(tarfile.is_tarfile(self.tarname))

# is_tarfile works on path-like objects

self.assertTrue(tarfile.is_tarfile(os_helper.FakePath(self.tarname)))

# is_tarfile works on file objects

with open(self.tarname, "rb") as fobj:

self.assertTrue(tarfile.is_tarfile(fobj))

# is_tarfile works on file-like objects

with open(self.tarname, "rb") as fobj:

self.assertTrue(tarfile.is_tarfile(io.BytesIO(fobj.read())))

def test_is_tarfile_keeps_position(self):

# Test for issue44289: tarfile.is_tarfile() modifies

# file object's current position

with open(self.tarname, "rb") as fobj:

tarfile.is_tarfile(fobj)

self.assertEqual(fobj.tell(), 0)

with open(self.tarname, "rb") as fobj:

file_like = io.BytesIO(fobj.read())

tarfile.is_tarfile(file_like)

self.assertEqual(file_like.tell(), 0)

def test_empty_tarfile(self):

# Test for issue6123: Allow opening empty archives.

# This test checks if tarfile.open() is able to open an empty tar

# archive successfully. Note that an empty tar archive is not the

# same as an empty file!

with tarfile.open(tmpname, self.mode.replace("r", "w")):

pass

try:

tar = tarfile.open(tmpname, self.mode)

tar.getnames()

except tarfile.ReadError:

self.fail("tarfile.open() failed on empty archive")

else:

self.assertListEqual(tar.getmembers(), [])

finally:

tar.close()

def test_non_existent_tarfile(self):

# Test for issue11513: prevent non-existent gzipped tarfiles raising

# multiple exceptions.

with self.assertRaisesRegex(FileNotFoundError, "xxx"):

tarfile.open("xxx", self.mode)

def test_null_tarfile(self):

# Test for issue6123: Allow opening empty archives.

# This test guarantees that tarfile.open() does not treat an empty

# file as an empty tar archive.

with open(tmpname, "wb"):

pass

self.assertRaises(tarfile.ReadError, tarfile.open, tmpname, self.mode)

self.assertRaises(tarfile.ReadError, tarfile.open, tmpname)

def test_ignore_zeros(self):

# Test TarFile's ignore_zeros option.

# generate 512 pseudorandom bytes

data = Random(0).randbytes(512)

for char in (b'\0', b'a'):

# Test if EOFHeaderError ('\0') and InvalidHeaderError ('a')

# are ignored correctly.

with self.open(tmpname, "w") as fobj:

fobj.write(char * 1024)

tarinfo = tarfile.TarInfo("foo")

tarinfo.size = len(data)

fobj.write(tarinfo.tobuf())

fobj.write(data)

tar = tarfile.open(tmpname, mode="r", ignore_zeros=True)

try:

self.assertListEqual(tar.getnames(), ["foo"],

"ignore_zeros=True should have skipped the %r-blocks" %

char)

finally:

tar.close()

def test_premature_end_of_archive(self):

for size in (512, 600, 1024, 1200):

with tarfile.open(tmpname, "w:") as tar:

t = tarfile.TarInfo("foo")

t.size = 1024

tar.addfile(t, io.BytesIO(b"a" * 1024))

with open(tmpname, "r+b") as fobj:

fobj.truncate(size)

with tarfile.open(tmpname) as tar:

with self.assertRaisesRegex(tarfile.ReadError, "unexpected end of data"):

for t in tar:

pass

with tarfile.open(tmpname) as tar:

t = tar.next()

with self.assertRaisesRegex(tarfile.ReadError, "unexpected end of data"):

tar.extract(t, TEMPDIR, filter='data')

with self.assertRaisesRegex(tarfile.ReadError, "unexpected end of data"):

tar.extractfile(t).read()

def test_length_zero_header(self):

# bpo-39017 (CVE-2019-20907): reading a zero-length header should fail

# with an exception

with self.assertRaisesRegex(tarfile.ReadError, "file could not be opened successfully"):

with tarfile.open(support.findfile('recursion.tar', subdir='archivetestdata')):

pass

def test_extractfile_attrs(self):

# gh-74468: TarFile.name must name a file, not a parent archive.

file = self.tar.getmember('ustar/regtype')

with self.tar.extractfile(file) as fobj:

self.assertEqual(fobj.name, 'ustar/regtype')

self.assertRaises(AttributeError, fobj.fileno)

self.assertEqual(fobj.mode, 'rb')

self.assertIs(fobj.readable(), True)

self.assertIs(fobj.writable(), False)

if self.is_stream:

self.assertRaises(AttributeError, fobj.seekable)

else:

self.assertIs(fobj.seekable(), True)

self.assertIs(fobj.closed, False)

self.assertIs(fobj.closed, True)

self.assertEqual(fobj.name, 'ustar/regtype')

self.assertRaises(AttributeError, fobj.fileno)

self.assertEqual(fobj.mode, 'rb')

self.assertIs(fobj.readable(), True)

self.assertIs(fobj.writable(), False)

if self.is_stream:

self.assertRaises(AttributeError, fobj.seekable)

else:

# Records the largest size ever passed to read(), so a test can check

# that tarfile does not request far more data than the archive holds

# (which on a real file would pre-allocate it).

def __init__(self, *args, **kwargs):

super().__init__(*args, **kwargs)

self.max_read_size = 0

def read(self, size=-1):

if size is not None and size >= 0:

self.max_read_size = max(self.max_read_size, size)

# gh-151497: the size of a GNU long name/link or a pax extended header is

# read from the archive and is untrusted. A crafted header can claim a

# size far larger than the file actually contains; opening such an archive

# must not try to read (and so pre-allocate) the claimed size in one go.

def crafted_archive(self, hdrtype):

tarinfo = tarfile.TarInfo("A")

tarinfo.type = hdrtype

tarinfo.size = 0xFFFFFFFF # ~4 GiB claimed in a 512-byte header

return tarinfo.tobuf(format=tarfile.GNU_FORMAT)

def check(self, hdrtype):

fobj = ReadSizeRecorder(self.crafted_archive(hdrtype))

try:

with tarfile.open(fileobj=fobj, mode="r:") as tar:

tar.getmembers()

except tarfile.ReadError:

pass # a truncated header is fine; we only check the allocation

# The bogus ~4 GiB size must never reach a single read() call.

self.assertLessEqual(fobj.max_read_size, tarfile._EXTHEADER_READ_CHUNK)

def test_gnu_longname_oversized_size(self):

self.check(tarfile.GNUTYPE_LONGNAME)

def test_gnu_longlink_oversized_size(self):

self.check(tarfile.GNUTYPE_LONGLINK)

def test_pax_header_oversized_size(self):

is_stream = False

def test_no_name_argument(self):

with open(self.tarname, "rb") as fobj:

self.assertIsInstance(fobj.name, str)

with tarfile.open(fileobj=fobj, mode=self.mode) as tar:

self.assertIsInstance(tar.name, str)

self.assertEqual(tar.name, os.path.abspath(fobj.name))

def test_no_name_attribute(self):

with open(self.tarname, "rb") as fobj:

data = fobj.read()

fobj = io.BytesIO(data)

self.assertRaises(AttributeError, getattr, fobj, "name")

tar = tarfile.open(fileobj=fobj, mode=self.mode)

self.assertIsNone(tar.name)

def test_empty_name_attribute(self):

with open(self.tarname, "rb") as fobj:

data = fobj.read()

fobj = io.BytesIO(data)

fobj.name = ""

with tarfile.open(fileobj=fobj, mode=self.mode) as tar:

self.assertIsNone(tar.name)

def test_int_name_attribute(self):

# Issue 21044: tarfile.open() should handle fileobj with an integer

# 'name' attribute.

fd = os.open(self.tarname, os.O_RDONLY)

with open(fd, 'rb') as fobj:

self.assertIsInstance(fobj.name, int)

with tarfile.open(fileobj=fobj, mode=self.mode) as tar:

self.assertIsNone(tar.name)

def test_bytes_name_attribute(self):

tarname = os.fsencode(self.tarname)

with open(tarname, 'rb') as fobj:

self.assertIsInstance(fobj.name, bytes)

with tarfile.open(fileobj=fobj, mode=self.mode) as tar:

self.assertIsInstance(tar.name, bytes)

self.assertEqual(tar.name, os.path.abspath(fobj.name))

def test_pathlike_name(self, tarname=None):

if tarname is None:

tarname = self.tarname

expected = os.path.abspath(tarname)

tarname = os_helper.FakePath(tarname)

with tarfile.open(tarname, mode=self.mode) as tar:

self.assertEqual(tar.name, expected)

with self.taropen(tarname) as tar:

self.assertEqual(tar.name, expected)

with tarfile.TarFile.open(tarname, mode=self.mode) as tar:

self.assertEqual(tar.name, expected)

if self.suffix == '':

with tarfile.TarFile(tarname, mode='r') as tar:

self.assertEqual(tar.name, expected)

def test_pathlike_bytes_name(self):

self.test_pathlike_name(os.fsencode(self.tarname))

def test_illegal_mode_arg(self):

with open(tmpname, 'wb'):

pass

with self.assertRaisesRegex(ValueError, 'mode must be '):

tar = self.taropen(tmpname, 'q')

with self.assertRaisesRegex(ValueError, 'mode must be '):

tar = self.taropen(tmpname, 'rw')

with self.assertRaisesRegex(ValueError, 'mode must be '):

tar = self.taropen(tmpname, '')

def test_fileobj_with_offset(self):

# Skip the first member and store values from the second member

# of the testtar.

tar = tarfile.open(self.tarname, mode=self.mode)

try:

tar.next()

t = tar.next()

name = t.name

offset = t.offset

with tar.extractfile(t) as f:

data = f.read()

finally:

tar.close()

# Open the testtar and seek to the offset of the second member.

with self.open(self.tarname) as fobj:

fobj.seek(offset)

# Test if the tarfile starts with the second member.

with tar.open(self.tarname, mode="r:", fileobj=fobj) as tar:

t = tar.next()

self.assertEqual(t.name, name)

# Read to the end of fileobj and test if seeking back to the

# beginning works.

tar.getmembers()

self.assertEqual(tar.extractfile(t).read(), data,

"seek back did not work")

def test_fail_comp(self):

# For Gzip and Bz2 Tests: fail with a ReadError on an uncompressed file.

self.assertRaises(tarfile.ReadError, tarfile.open, tarname, self.mode)

with open(tarname, "rb") as fobj:

self.assertRaises(tarfile.ReadError, tarfile.open,

fileobj=fobj, mode=self.mode)

def test_v7_dirtype(self):

# Test old style dirtype member (bug #1336623):

# Old V7 tars create directory members using an AREGTYPE

# header with a "/" appended to the filename field.

tarinfo = self.tar.getmember("misc/dirtype-old-v7")

self.assertEqual(tarinfo.type, tarfile.DIRTYPE,

"v7 dirtype failed")

def test_xstar_type(self):

# The xstar format stores extra atime and ctime fields inside the

# space reserved for the prefix field. The prefix field must be

# ignored in this case, otherwise it will mess up the name.

try:

self.tar.getmember("misc/regtype-xstar")

except KeyError:

self.fail("failed to find misc/regtype-xstar (mangled prefix?)")

def test_check_members(self):

for tarinfo in self.tar:

self.assertEqual(int(tarinfo.mtime), 0o7606136617,

"wrong mtime for %s" % tarinfo.name)

if not tarinfo.name.startswith("ustar/"):

continue

self.assertEqual(tarinfo.uname, "tarfile",

"wrong uname for %s" % tarinfo.name)

def test_find_members(self):

self.assertEqual(self.tar.getmembers()[-1].name, "misc/eof",

"could not find all members")

@unittest.skipUnless(hasattr(os, "link"),

@os_helper.skip_unless_symlink

def test_extract_hardlink(self):

# Test hardlink extraction (e.g. bug #857297).

with tarfile.open(tarname, errorlevel=1, encoding="iso8859-1") as tar:

tar.extract("ustar/regtype", TEMPDIR, filter='data')

self.addCleanup(os_helper.unlink, os.path.join(TEMPDIR, "ustar/regtype"))

tar.extract("ustar/lnktype", TEMPDIR, filter='data')

self.addCleanup(os_helper.unlink, os.path.join(TEMPDIR, "ustar/lnktype"))

with open(os.path.join(TEMPDIR, "ustar/lnktype"), "rb") as f:

data = f.read()

self.assertEqual(sha256sum(data), sha256_regtype)

tar.extract("ustar/symtype", TEMPDIR, filter='data')

self.addCleanup(os_helper.unlink, os.path.join(TEMPDIR, "ustar/symtype"))

with open(os.path.join(TEMPDIR, "ustar/symtype"), "rb") as f:

data = f.read()

self.assertEqual(sha256sum(data), sha256_regtype)

@os_helper.skip_unless_working_chmod

def test_extractall(self):

# Test if extractall() correctly restores directory permissions

# and times (see issue1735).

tar = tarfile.open(tarname, encoding="iso8859-1")

DIR = os.path.join(TEMPDIR, "extractall")

os.mkdir(DIR)

try:

directories = [t for t in tar if t.isdir()]

tar.extractall(DIR, directories, filter='fully_trusted')

for tarinfo in directories:

path = os.path.join(DIR, tarinfo.name)

if sys.platform != "win32":

# Win32 has no support for fine grained permissions.

self.assertEqual(tarinfo.mode & 0o777,

os.stat(path).st_mode & 0o777,

tarinfo.name)

def format_mtime(mtime):

if isinstance(mtime, float):

return "{} ({})".format(mtime, mtime.hex())

else:

return "{!r} (int)".format(mtime)

file_mtime = os.path.getmtime(path)

errmsg = "tar mtime {0} != file time {1} of path {2!a}".format(

format_mtime(tarinfo.mtime),

format_mtime(file_mtime),

path)

self.assertEqual(tarinfo.mtime, file_mtime, errmsg)

finally:

tar.close()

os_helper.rmtree(DIR)

@staticmethod

def test_extractall_default_filter():

# Test that the default filter is now "data", and the other filter types are not used.

DIR = pathlib.Path(TEMPDIR) / "extractall_default_filter"

with (

os_helper.temp_dir(DIR),

tarfile.open(tarname, encoding="iso8859-1") as tar,

unittest.mock.patch("tarfile.data_filter", wraps=tarfile.data_filter) as mock_data_filter,

unittest.mock.patch("tarfile.tar_filter", wraps=tarfile.tar_filter) as mock_tar_filter,

unittest.mock.patch("tarfile.fully_trusted_filter", wraps=tarfile.fully_trusted_filter) as mock_ft_filter

):

directories = [t for t in tar if t.isdir()]

tar.extractall(DIR, directories)

mock_data_filter.assert_called()

mock_ft_filter.assert_not_called()

mock_tar_filter.assert_not_called()

@os_helper.skip_unless_working_chmod

def test_extract_directory(self):

dirtype = "ustar/dirtype"

DIR = os.path.join(TEMPDIR, "extractdir")

os.mkdir(DIR)

try:

with tarfile.open(tarname, encoding="iso8859-1") as tar:

tarinfo = tar.getmember(dirtype)

tar.extract(tarinfo, path=DIR, filter='fully_trusted')

extracted = os.path.join(DIR, dirtype)

self.assertEqual(os.path.getmtime(extracted), tarinfo.mtime)

if sys.platform != "win32":

self.assertEqual(os.stat(extracted).st_mode & 0o777, 0o755)

finally:

os_helper.rmtree(DIR)

def test_extractall_pathlike_dir(self):

DIR = os.path.join(TEMPDIR, "extractall")

with os_helper.temp_dir(DIR), \

tarfile.open(tarname, encoding="iso8859-1") as tar:

directories = [t for t in tar if t.isdir()]

tar.extractall(os_helper.FakePath(DIR), directories, filter='fully_trusted')

for tarinfo in directories:

path = os.path.join(DIR, tarinfo.name)

self.assertEqual(os.path.getmtime(path), tarinfo.mtime)

def test_extract_pathlike_dir(self):

dirtype = "ustar/dirtype"

DIR = os.path.join(TEMPDIR, "extractall")

with os_helper.temp_dir(DIR), \

tarfile.open(tarname, encoding="iso8859-1") as tar:

tarinfo = tar.getmember(dirtype)

tar.extract(tarinfo, path=os_helper.FakePath(DIR), filter='fully_trusted')

extracted = os.path.join(DIR, dirtype)

self.assertEqual(os.path.getmtime(extracted), tarinfo.mtime)

def test_init_close_fobj(self):

# Issue #7341: Close the internal file object in the TarFile

# constructor in case of an error. For the test we rely on

# the fact that opening an empty file raises a ReadError.

empty = os.path.join(TEMPDIR, "empty")

with open(empty, "wb") as fobj:

fobj.write(b"")

try:

tar = object.__new__(tarfile.TarFile)

try:

tar.__init__(empty)

except tarfile.ReadError:

self.assertTrue(tar.fileobj.closed)

else:

self.fail("ReadError not raised")

finally:

os_helper.unlink(empty)

def test_parallel_iteration(self):

# Issue #16601: Restarting iteration over tarfile continued

# from where it left off.

with tarfile.open(self.tarname) as tar:

for m1, m2 in zip(tar, tar):

self.assertEqual(m1.offset, m2.offset)

self.assertEqual(m1.get_info(), m2.get_info())

@unittest.skipIf(zlib is None, "requires zlib")

def test_zlib_error_does_not_leak(self):

# bpo-39039: tarfile.open allowed zlib exceptions to bubble up when

# parsing certain types of invalid data

with unittest.mock.patch("tarfile.TarInfo.fromtarfile") as mock:

mock.side_effect = zlib.error

with self.assertRaises(tarfile.ReadError):

tarfile.open(self.tarname)

def test_next_on_empty_tarfile(self):

fd = io.BytesIO()

tf = tarfile.open(fileobj=fd, mode="w")

tf.close()

fd.seek(0)

with tarfile.open(fileobj=fd, mode="r|") as tf:

self.assertEqual(tf.next(), None)

fd.seek(0)

with tarfile.open(fileobj=fd, mode="r") as tf:

self.assertEqual(tf.next(), None)

def _setup_symlink_to_target(self, temp_dirpath):

target_filepath = os.path.join(temp_dirpath, "target")

ustar_dirpath = os.path.join(temp_dirpath, "ustar")

hardlink_filepath = os.path.join(ustar_dirpath, "lnktype")

with open(target_filepath, "wb") as f:

f.write(b"target")

os.makedirs(ustar_dirpath)

os.symlink(target_filepath, hardlink_filepath)

return target_filepath, hardlink_filepath

def _assert_on_file_content(self, filepath, digest):

with open(filepath, "rb") as f:

data = f.read()

self.assertEqual(sha256sum(data), digest)

@unittest.skipUnless(

@os_helper.skip_unless_symlink

def test_extract_hardlink_on_symlink(self):

"""

with os_helper.temp_dir() as DIR:

target_filepath, hardlink_filepath = self._setup_symlink_to_target(DIR)

with tarfile.open(tarname, encoding="iso8859-1") as tar:

tar.extract("ustar/regtype", DIR, filter="data")

tar.extract("ustar/lnktype", DIR, filter="data")

self._assert_on_file_content(target_filepath, sha256sum(b"target"))

self._assert_on_file_content(hardlink_filepath, sha256_regtype)

@unittest.skipUnless(

@os_helper.skip_unless_symlink

def test_extractall_hardlink_on_symlink(self):

"""

with os_helper.temp_dir() as DIR:

target_filepath, hardlink_filepath = self._setup_symlink_to_target(DIR)

with tarfile.open(tarname, encoding="iso8859-1") as tar:

tar.extractall(

DIR, members=["ustar/regtype", "ustar/lnktype"], filter="data",

)

self._assert_on_file_content(target_filepath, sha256sum(b"target"))

def test_read_with_extra_field(self):

with open(self.tarname, 'rb') as f:

data = bytearray(f.read())

flags = data[3]

self.assertEqual(flags, 8)

data[3] = flags | 4

data[10:10] = b'\x05\x00extra'

with open(tmpname, 'wb') as f:

f.write(data)

print(self.mode)

with tarfile.open(tmpname, mode=self.mode):

pass

def test_read_with_file_comment(self):

with open(self.tarname, 'rb') as f:

data = bytearray(f.read())

flags = data[3]

self.assertEqual(flags, 8)

data[3] = flags | 16

i = data.index(0, 10) + 1

data[i:i] = b'comment\x00'

with open(tmpname, 'wb') as f:

f.write(data)

with tarfile.open(tmpname, mode=self.mode):

prefix="r|"