diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml index 3ec7934..e28832c 100644 --- a/.github/workflows/test.yml +++ b/.github/workflows/test.yml @@ -9,14 +9,15 @@ on: jobs: tests: strategy: + max-parallel: 3 matrix: - python-version: [3.8, 3.11, 3.12] - os: [ubuntu-latest, macos-latest, windows-latest] + python-version: ['3.8', '3.9', '3.10', '3.11', '3.12'] + os: [ubuntu-latest, windows-latest, macos-latest] runs-on: ${{ matrix.os }} steps: - - uses: actions/checkout@v3 + - uses: actions/checkout@v4 - name: Set up Python ${{ matrix.python-version }} - uses: actions/setup-python@v4 + uses: actions/setup-python@v5 with: python-version: ${{ matrix.python-version }} - name: Install zipfile2 @@ -25,12 +26,16 @@ jobs: run: python -m pip install -r "dev_requirements.txt" - name: Test with haas run: python -m haas zipfile2 + - name: Integration tests + run: python -m haas . + working-directory: cpython-tests/${{ matrix.python-version }} + code-lint: runs-on: ubuntu-latest steps: - - uses: actions/checkout@v3 + - uses: actions/checkout@v4 - name: Set up Python 3.8 - uses: actions/setup-python@v4 + uses: actions/setup-python@v5 with: python-version: 3.8 - name: Install flake8 diff --git a/.gitignore b/.gitignore index 3823f17..165a842 100644 --- a/.gitignore +++ b/.gitignore @@ -2,3 +2,4 @@ build .tox zipfile2/_version.py zipfile2*egg-info +__pycache__ \ No newline at end of file diff --git a/cpython-tests/3.10/test_zipfile.py b/cpython-tests/3.10/test_zipfile.py new file mode 100644 index 0000000..c527101 --- /dev/null +++ b/cpython-tests/3.10/test_zipfile.py @@ -0,0 +1,3432 @@ +import array +import contextlib +import importlib.util +import io +import itertools +import os +import pathlib +import posixpath +import string +import struct +import subprocess +import sys +from test.support.script_helper import assert_python_ok +import time +import unittest +import unittest.mock as mock +import zipfile +import functools + + +from tempfile import TemporaryFile +from random import randint, random, randbytes + +from test.support import script_helper +from test.support import (findfile, requires_zlib, requires_bz2, + requires_lzma, captured_stdout) +from test.support.os_helper import TESTFN, unlink, rmtree, temp_dir, temp_cwd + +import zipfile2 + + +TESTFN2 = TESTFN + "2" +TESTFNDIR = TESTFN + "d" +FIXEDTEST_SIZE = 1000 +DATAFILES_DIR = 'zipfile_datafiles' + +SMALL_TEST_DATA = [('_ziptest1', '1q2w3e4r5t'), + ('ziptest2dir/_ziptest2', 'qawsedrftg'), + ('ziptest2dir/ziptest3dir/_ziptest3', 'azsxdcfvgb'), + ('ziptest2dir/ziptest3dir/ziptest4dir/_ziptest3', '6y7u8i9o0p')] + +def get_files(test): + yield TESTFN2 + with TemporaryFile() as f: + yield f + test.assertFalse(f.closed) + with io.BytesIO() as f: + yield f + test.assertFalse(f.closed) + +class AbstractTestsWithSourceFile: + @classmethod + def setUpClass(cls): + cls.line_gen = [bytes("Zipfile test line %d. random float: %f\n" % + (i, random()), "ascii") + for i in range(FIXEDTEST_SIZE)] + cls.data = b''.join(cls.line_gen) + + def setUp(self): + # Make a source file with some lines + with open(TESTFN, "wb") as fp: + fp.write(self.data) + + def make_test_archive(self, f, compression, compresslevel=None): + kwargs = {'compression': compression, 'compresslevel': compresslevel} + # Create the ZIP archive + with zipfile2.ZipFile(f, "w", **kwargs) as zipfp: + zipfp.write(TESTFN, "another.name") + zipfp.write(TESTFN, TESTFN) + zipfp.writestr("strfile", self.data) + with zipfp.open('written-open-w', mode='w') as f: + for line in self.line_gen: + f.write(line) + + def zip_test(self, f, compression, compresslevel=None): + self.make_test_archive(f, compression, compresslevel) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + self.assertEqual(zipfp.read(TESTFN), self.data) + self.assertEqual(zipfp.read("another.name"), self.data) + self.assertEqual(zipfp.read("strfile"), self.data) + + # Print the ZIP directory + fp = io.StringIO() + zipfp.printdir(file=fp) + directory = fp.getvalue() + lines = directory.splitlines() + self.assertEqual(len(lines), 5) # Number of files + header + + self.assertIn('File Name', lines[0]) + self.assertIn('Modified', lines[0]) + self.assertIn('Size', lines[0]) + + fn, date, time_, size = lines[1].split() + self.assertEqual(fn, 'another.name') + self.assertTrue(time.strptime(date, '%Y-%m-%d')) + self.assertTrue(time.strptime(time_, '%H:%M:%S')) + self.assertEqual(size, str(len(self.data))) + + # Check the namelist + names = zipfp.namelist() + self.assertEqual(len(names), 4) + self.assertIn(TESTFN, names) + self.assertIn("another.name", names) + self.assertIn("strfile", names) + self.assertIn("written-open-w", names) + + # Check infolist + infos = zipfp.infolist() + names = [i.filename for i in infos] + self.assertEqual(len(names), 4) + self.assertIn(TESTFN, names) + self.assertIn("another.name", names) + self.assertIn("strfile", names) + self.assertIn("written-open-w", names) + for i in infos: + self.assertEqual(i.file_size, len(self.data)) + + # check getinfo + for nm in (TESTFN, "another.name", "strfile", "written-open-w"): + info = zipfp.getinfo(nm) + self.assertEqual(info.filename, nm) + self.assertEqual(info.file_size, len(self.data)) + + # Check that testzip doesn't raise an exception + zipfp.testzip() + + def test_basic(self): + for f in get_files(self): + self.zip_test(f, self.compression) + + def zip_open_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + zipdata1 = [] + with zipfp.open(TESTFN) as zipopen1: + while True: + read_data = zipopen1.read(256) + if not read_data: + break + zipdata1.append(read_data) + + zipdata2 = [] + with zipfp.open("another.name") as zipopen2: + while True: + read_data = zipopen2.read(256) + if not read_data: + break + zipdata2.append(read_data) + + self.assertEqual(b''.join(zipdata1), self.data) + self.assertEqual(b''.join(zipdata2), self.data) + + def test_open(self): + for f in get_files(self): + self.zip_open_test(f, self.compression) + + def test_open_with_pathlike(self): + path = pathlib.Path(TESTFN2) + self.zip_open_test(path, self.compression) + with zipfile2.ZipFile(path, "r", self.compression) as zipfp: + self.assertIsInstance(zipfp.filename, str) + + def zip_random_open_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + zipdata1 = [] + with zipfp.open(TESTFN) as zipopen1: + while True: + read_data = zipopen1.read(randint(1, 1024)) + if not read_data: + break + zipdata1.append(read_data) + + self.assertEqual(b''.join(zipdata1), self.data) + + def test_random_open(self): + for f in get_files(self): + self.zip_random_open_test(f, self.compression) + + def zip_read1_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp, \ + zipfp.open(TESTFN) as zipopen: + zipdata = [] + while True: + read_data = zipopen.read1(-1) + if not read_data: + break + zipdata.append(read_data) + + self.assertEqual(b''.join(zipdata), self.data) + + def test_read1(self): + for f in get_files(self): + self.zip_read1_test(f, self.compression) + + def zip_read1_10_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp, \ + zipfp.open(TESTFN) as zipopen: + zipdata = [] + while True: + read_data = zipopen.read1(10) + self.assertLessEqual(len(read_data), 10) + if not read_data: + break + zipdata.append(read_data) + + self.assertEqual(b''.join(zipdata), self.data) + + def test_read1_10(self): + for f in get_files(self): + self.zip_read1_10_test(f, self.compression) + + def zip_readline_read_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp, \ + zipfp.open(TESTFN) as zipopen: + data = b'' + while True: + read = zipopen.readline() + if not read: + break + data += read + + read = zipopen.read(100) + if not read: + break + data += read + + self.assertEqual(data, self.data) + + def test_readline_read(self): + # Issue #7610: calls to readline() interleaved with calls to read(). + for f in get_files(self): + self.zip_readline_read_test(f, self.compression) + + def zip_readline_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp: + with zipfp.open(TESTFN) as zipopen: + for line in self.line_gen: + linedata = zipopen.readline() + self.assertEqual(linedata, line) + + def test_readline(self): + for f in get_files(self): + self.zip_readline_test(f, self.compression) + + def zip_readlines_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp: + with zipfp.open(TESTFN) as zipopen: + ziplines = zipopen.readlines() + for line, zipline in zip(self.line_gen, ziplines): + self.assertEqual(zipline, line) + + def test_readlines(self): + for f in get_files(self): + self.zip_readlines_test(f, self.compression) + + def zip_iterlines_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp: + with zipfp.open(TESTFN) as zipopen: + for line, zipline in zip(self.line_gen, zipopen): + self.assertEqual(zipline, line) + + def test_iterlines(self): + for f in get_files(self): + self.zip_iterlines_test(f, self.compression) + + def test_low_compression(self): + """Check for cases where compressed data is larger than original.""" + # Create the ZIP archive + with zipfile2.ZipFile(TESTFN2, "w", self.compression) as zipfp: + zipfp.writestr("strfile", '12') + + # Get an open object for strfile + with zipfile2.ZipFile(TESTFN2, "r", self.compression) as zipfp: + with zipfp.open("strfile") as openobj: + self.assertEqual(openobj.read(1), b'1') + self.assertEqual(openobj.read(1), b'2') + + def test_writestr_compression(self): + zipfp = zipfile2.ZipFile(TESTFN2, "w") + zipfp.writestr("b.txt", "hello world", compress_type=self.compression) + info = zipfp.getinfo('b.txt') + self.assertEqual(info.compress_type, self.compression) + + def test_writestr_compresslevel(self): + zipfp = zipfile2.ZipFile(TESTFN2, "w", compresslevel=1) + zipfp.writestr("a.txt", "hello world", compress_type=self.compression) + zipfp.writestr("b.txt", "hello world", compress_type=self.compression, + compresslevel=2) + + # Compression level follows the constructor. + a_info = zipfp.getinfo('a.txt') + self.assertEqual(a_info.compress_type, self.compression) + self.assertEqual(a_info._compresslevel, 1) + + # Compression level is overridden. + b_info = zipfp.getinfo('b.txt') + self.assertEqual(b_info.compress_type, self.compression) + self.assertEqual(b_info._compresslevel, 2) + + def test_read_return_size(self): + # Issue #9837: ZipExtFile.read() shouldn't return more bytes + # than requested. + for test_size in (1, 4095, 4096, 4097, 16384): + file_size = test_size + 1 + junk = randbytes(file_size) + with zipfile2.ZipFile(io.BytesIO(), "w", self.compression) as zipf: + zipf.writestr('foo', junk) + with zipf.open('foo', 'r') as fp: + buf = fp.read(test_size) + self.assertEqual(len(buf), test_size) + + def test_truncated_zipfile(self): + fp = io.BytesIO() + with zipfile2.ZipFile(fp, mode='w') as zipf: + zipf.writestr('strfile', self.data, compress_type=self.compression) + end_offset = fp.tell() + zipfiledata = fp.getvalue() + + fp = io.BytesIO(zipfiledata) + with zipfile2.ZipFile(fp) as zipf: + with zipf.open('strfile') as zipopen: + fp.truncate(end_offset - 20) + with self.assertRaises(EOFError): + zipopen.read() + + fp = io.BytesIO(zipfiledata) + with zipfile2.ZipFile(fp) as zipf: + with zipf.open('strfile') as zipopen: + fp.truncate(end_offset - 20) + with self.assertRaises(EOFError): + while zipopen.read(100): + pass + + fp = io.BytesIO(zipfiledata) + with zipfile2.ZipFile(fp) as zipf: + with zipf.open('strfile') as zipopen: + fp.truncate(end_offset - 20) + with self.assertRaises(EOFError): + while zipopen.read1(100): + pass + + def test_repr(self): + fname = 'file.name' + for f in get_files(self): + with zipfile2.ZipFile(f, 'w', self.compression) as zipfp: + zipfp.write(TESTFN, fname) + r = repr(zipfp) + self.assertIn("mode='w'", r) + + with zipfile2.ZipFile(f, 'r') as zipfp: + r = repr(zipfp) + if isinstance(f, str): + self.assertIn('filename=%r' % f, r) + else: + self.assertIn('file=%r' % f, r) + self.assertIn("mode='r'", r) + r = repr(zipfp.getinfo(fname)) + self.assertIn('filename=%r' % fname, r) + self.assertIn('filemode=', r) + self.assertIn('file_size=', r) + if self.compression != zipfile.ZIP_STORED: + self.assertIn('compress_type=', r) + self.assertIn('compress_size=', r) + with zipfp.open(fname) as zipopen: + r = repr(zipopen) + self.assertIn('name=%r' % fname, r) + self.assertIn("mode='r'", r) + if self.compression != zipfile.ZIP_STORED: + self.assertIn('compress_type=', r) + self.assertIn('[closed]', repr(zipopen)) + self.assertIn('[closed]', repr(zipfp)) + + def test_compresslevel_basic(self): + for f in get_files(self): + self.zip_test(f, self.compression, compresslevel=9) + + def test_per_file_compresslevel(self): + """Check that files within a Zip archive can have different + compression levels.""" + with zipfile2.ZipFile(TESTFN2, "w", compresslevel=1) as zipfp: + zipfp.write(TESTFN, 'compress_1') + zipfp.write(TESTFN, 'compress_9', compresslevel=9) + one_info = zipfp.getinfo('compress_1') + nine_info = zipfp.getinfo('compress_9') + self.assertEqual(one_info._compresslevel, 1) + self.assertEqual(nine_info._compresslevel, 9) + + def test_writing_errors(self): + class BrokenFile(io.BytesIO): + def write(self, data): + nonlocal count + if count is not None: + if count == stop: + raise OSError + count += 1 + super().write(data) + + stop = 0 + while True: + testfile = BrokenFile() + count = None + with zipfile2.ZipFile(testfile, 'w', self.compression) as zipfp: + with zipfp.open('file1', 'w') as f: + f.write(b'data1') + count = 0 + try: + with zipfp.open('file2', 'w') as f: + f.write(b'data2') + except OSError: + stop += 1 + else: + break + finally: + count = None + with zipfile2.ZipFile(io.BytesIO(testfile.getvalue())) as zipfp: + self.assertEqual(zipfp.namelist(), ['file1']) + self.assertEqual(zipfp.read('file1'), b'data1') + + with zipfile2.ZipFile(io.BytesIO(testfile.getvalue())) as zipfp: + self.assertEqual(zipfp.namelist(), ['file1', 'file2']) + self.assertEqual(zipfp.read('file1'), b'data1') + self.assertEqual(zipfp.read('file2'), b'data2') + + + def tearDown(self): + unlink(TESTFN) + unlink(TESTFN2) + + +class StoredTestsWithSourceFile(AbstractTestsWithSourceFile, + unittest.TestCase): + compression = zipfile.ZIP_STORED + test_low_compression = None + + def zip_test_writestr_permissions(self, f, compression): + # Make sure that writestr and open(... mode='w') create files with + # mode 0600, when they are passed a name rather than a ZipInfo + # instance. + + self.make_test_archive(f, compression) + with zipfile2.ZipFile(f, "r") as zipfp: + zinfo = zipfp.getinfo('strfile') + self.assertEqual(zinfo.external_attr, 0o600 << 16) + + zinfo2 = zipfp.getinfo('written-open-w') + self.assertEqual(zinfo2.external_attr, 0o600 << 16) + + def test_writestr_permissions(self): + for f in get_files(self): + self.zip_test_writestr_permissions(f, zipfile.ZIP_STORED) + + def test_absolute_arcnames(self): + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, "/absolute") + + with zipfile2.ZipFile(TESTFN2, "r", zipfile.ZIP_STORED) as zipfp: + self.assertEqual(zipfp.namelist(), ["absolute"]) + + def test_append_to_zip_file(self): + """Test appending to an existing zipfile.""" + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + + with zipfile2.ZipFile(TESTFN2, "a", zipfile.ZIP_STORED) as zipfp: + zipfp.writestr("strfile", self.data) + self.assertEqual(zipfp.namelist(), [TESTFN, "strfile"]) + + def test_append_to_non_zip_file(self): + """Test appending to an existing file that is not a zipfile.""" + # NOTE: this test fails if len(d) < 22 because of the first + # line "fpin.seek(-22, 2)" in _EndRecData + data = b'I am not a ZipFile!'*10 + with open(TESTFN2, 'wb') as f: + f.write(data) + + with zipfile2.ZipFile(TESTFN2, "a", zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + + with open(TESTFN2, 'rb') as f: + f.seek(len(data)) + with zipfile2.ZipFile(f, "r") as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN]) + self.assertEqual(zipfp.read(TESTFN), self.data) + with open(TESTFN2, 'rb') as f: + self.assertEqual(f.read(len(data)), data) + zipfiledata = f.read() + with io.BytesIO(zipfiledata) as bio, zipfile2.ZipFile(bio) as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN]) + self.assertEqual(zipfp.read(TESTFN), self.data) + + def test_read_concatenated_zip_file(self): + with io.BytesIO() as bio: + with zipfile2.ZipFile(bio, 'w', zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + zipfiledata = bio.getvalue() + data = b'I am not a ZipFile!'*10 + with open(TESTFN2, 'wb') as f: + f.write(data) + f.write(zipfiledata) + + with zipfile2.ZipFile(TESTFN2) as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN]) + self.assertEqual(zipfp.read(TESTFN), self.data) + + def test_append_to_concatenated_zip_file(self): + with io.BytesIO() as bio: + with zipfile2.ZipFile(bio, 'w', zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + zipfiledata = bio.getvalue() + data = b'I am not a ZipFile!'*1000000 + with open(TESTFN2, 'wb') as f: + f.write(data) + f.write(zipfiledata) + + with zipfile2.ZipFile(TESTFN2, 'a') as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN]) + zipfp.writestr('strfile', self.data) + + with open(TESTFN2, 'rb') as f: + self.assertEqual(f.read(len(data)), data) + zipfiledata = f.read() + with io.BytesIO(zipfiledata) as bio, zipfile2.ZipFile(bio) as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN, 'strfile']) + self.assertEqual(zipfp.read(TESTFN), self.data) + self.assertEqual(zipfp.read('strfile'), self.data) + + def test_ignores_newline_at_end(self): + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + with open(TESTFN2, 'a', encoding='utf-8') as f: + f.write("\r\n\00\00\00") + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertIsInstance(zipfp, zipfile2.ZipFile) + + def test_ignores_stuff_appended_past_comments(self): + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED) as zipfp: + zipfp.comment = b"this is a comment" + zipfp.write(TESTFN, TESTFN) + with open(TESTFN2, 'a', encoding='utf-8') as f: + f.write("abcdef\r\n") + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertIsInstance(zipfp, zipfile2.ZipFile) + self.assertEqual(zipfp.comment, b"this is a comment") + + def test_write_default_name(self): + """Check that calling ZipFile.write without arcname specified + produces the expected result.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + zipfp.write(TESTFN) + with open(TESTFN, "rb") as f: + self.assertEqual(zipfp.read(TESTFN), f.read()) + + def test_io_on_closed_zipextfile(self): + fname = "somefile.txt" + with zipfile2.ZipFile(TESTFN2, mode="w") as zipfp: + zipfp.writestr(fname, "bogus") + + with zipfile2.ZipFile(TESTFN2, mode="r") as zipfp: + with zipfp.open(fname) as fid: + fid.close() + self.assertRaises(ValueError, fid.read) + self.assertRaises(ValueError, fid.seek, 0) + self.assertRaises(ValueError, fid.tell) + self.assertRaises(ValueError, fid.readable) + self.assertRaises(ValueError, fid.seekable) + + def test_write_to_readonly(self): + """Check that trying to call write() on a readonly ZipFile object + raises a ValueError.""" + with zipfile2.ZipFile(TESTFN2, mode="w") as zipfp: + zipfp.writestr("somefile.txt", "bogus") + + with zipfile2.ZipFile(TESTFN2, mode="r") as zipfp: + self.assertRaises(ValueError, zipfp.write, TESTFN) + + with zipfile2.ZipFile(TESTFN2, mode="r") as zipfp: + with self.assertRaises(ValueError): + zipfp.open(TESTFN, mode='w') + + def test_add_file_before_1980(self): + # Set atime and mtime to 1970-01-01 + os.utime(TESTFN, (0, 0)) + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + self.assertRaises(ValueError, zipfp.write, TESTFN) + + with zipfile2.ZipFile(TESTFN2, "w", strict_timestamps=False) as zipfp: + zipfp.write(TESTFN) + zinfo = zipfp.getinfo(TESTFN) + self.assertEqual(zinfo.date_time, (1980, 1, 1, 0, 0, 0)) + + def test_add_file_after_2107(self): + # Set atime and mtime to 2108-12-30 + ts = 4386268800 + try: + time.localtime(ts) + except OverflowError: + self.skipTest(f'time.localtime({ts}) raises OverflowError') + try: + os.utime(TESTFN, (ts, ts)) + except OverflowError: + self.skipTest('Host fs cannot set timestamp to required value.') + + mtime_ns = os.stat(TESTFN).st_mtime_ns + if mtime_ns != (4386268800 * 10**9): + # XFS filesystem is limited to 32-bit timestamp, but the syscall + # didn't fail. Moreover, there is a VFS bug which returns + # a cached timestamp which is different than the value on disk. + # + # Test st_mtime_ns rather than st_mtime to avoid rounding issues. + # + # https://bugzilla.redhat.com/show_bug.cgi?id=1795576 + # https://bugs.python.org/issue39460#msg360952 + self.skipTest(f"Linux VFS/XFS kernel bug detected: {mtime_ns=}") + + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + self.assertRaises(struct.error, zipfp.write, TESTFN) + + with zipfile2.ZipFile(TESTFN2, "w", strict_timestamps=False) as zipfp: + zipfp.write(TESTFN) + zinfo = zipfp.getinfo(TESTFN) + self.assertEqual(zinfo.date_time, (2107, 12, 31, 23, 59, 59)) + + +@requires_zlib() +class DeflateTestsWithSourceFile(AbstractTestsWithSourceFile, + unittest.TestCase): + compression = zipfile.ZIP_DEFLATED + + def test_per_file_compression(self): + """Check that files within a Zip archive can have different + compression options.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + zipfp.write(TESTFN, 'storeme', zipfile.ZIP_STORED) + zipfp.write(TESTFN, 'deflateme', zipfile.ZIP_DEFLATED) + sinfo = zipfp.getinfo('storeme') + dinfo = zipfp.getinfo('deflateme') + self.assertEqual(sinfo.compress_type, zipfile.ZIP_STORED) + self.assertEqual(dinfo.compress_type, zipfile.ZIP_DEFLATED) + +@requires_bz2() +class Bzip2TestsWithSourceFile(AbstractTestsWithSourceFile, + unittest.TestCase): + compression = zipfile.ZIP_BZIP2 + +@requires_lzma() +class LzmaTestsWithSourceFile(AbstractTestsWithSourceFile, + unittest.TestCase): + compression = zipfile.ZIP_LZMA + + +class AbstractTestZip64InSmallFiles: + # These tests test the ZIP64 functionality without using large files, + # see test_zipfile64 for proper tests. + + @classmethod + def setUpClass(cls): + line_gen = (bytes("Test of zipfile line %d." % i, "ascii") + for i in range(0, FIXEDTEST_SIZE)) + cls.data = b'\n'.join(line_gen) + + def setUp(self): + self._limit = zipfile.ZIP64_LIMIT + self._filecount_limit = zipfile.ZIP_FILECOUNT_LIMIT + zipfile.ZIP64_LIMIT = 1000 + zipfile.ZIP_FILECOUNT_LIMIT = 9 + + # Make a source file with some lines + with open(TESTFN, "wb") as fp: + fp.write(self.data) + + def zip_test(self, f, compression): + # Create the ZIP archive + with zipfile2.ZipFile(f, "w", compression, allowZip64=True) as zipfp: + zipfp.write(TESTFN, "another.name") + zipfp.write(TESTFN, TESTFN) + zipfp.writestr("strfile", self.data) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + self.assertEqual(zipfp.read(TESTFN), self.data) + self.assertEqual(zipfp.read("another.name"), self.data) + self.assertEqual(zipfp.read("strfile"), self.data) + + # Print the ZIP directory + fp = io.StringIO() + zipfp.printdir(fp) + + directory = fp.getvalue() + lines = directory.splitlines() + self.assertEqual(len(lines), 4) # Number of files + header + + self.assertIn('File Name', lines[0]) + self.assertIn('Modified', lines[0]) + self.assertIn('Size', lines[0]) + + fn, date, time_, size = lines[1].split() + self.assertEqual(fn, 'another.name') + self.assertTrue(time.strptime(date, '%Y-%m-%d')) + self.assertTrue(time.strptime(time_, '%H:%M:%S')) + self.assertEqual(size, str(len(self.data))) + + # Check the namelist + names = zipfp.namelist() + self.assertEqual(len(names), 3) + self.assertIn(TESTFN, names) + self.assertIn("another.name", names) + self.assertIn("strfile", names) + + # Check infolist + infos = zipfp.infolist() + names = [i.filename for i in infos] + self.assertEqual(len(names), 3) + self.assertIn(TESTFN, names) + self.assertIn("another.name", names) + self.assertIn("strfile", names) + for i in infos: + self.assertEqual(i.file_size, len(self.data)) + + # check getinfo + for nm in (TESTFN, "another.name", "strfile"): + info = zipfp.getinfo(nm) + self.assertEqual(info.filename, nm) + self.assertEqual(info.file_size, len(self.data)) + + # Check that testzip doesn't raise an exception + zipfp.testzip() + + def test_basic(self): + for f in get_files(self): + self.zip_test(f, self.compression) + + def test_too_many_files(self): + # This test checks that more than 64k files can be added to an archive, + # and that the resulting archive can be read properly by ZipFile + zipf = zipfile2.ZipFile(TESTFN, "w", self.compression, + allowZip64=True) + zipf.debug = 100 + numfiles = 15 + for i in range(numfiles): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles) + zipf.close() + + zipf2 = zipfile2.ZipFile(TESTFN, "r", self.compression) + self.assertEqual(len(zipf2.namelist()), numfiles) + for i in range(numfiles): + content = zipf2.read("foo%08d" % i).decode('ascii') + self.assertEqual(content, "%d" % (i**3 % 57)) + zipf2.close() + + def test_too_many_files_append(self): + zipf = zipfile2.ZipFile(TESTFN, "w", self.compression, + allowZip64=False) + zipf.debug = 100 + numfiles = 9 + for i in range(numfiles): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles) + with self.assertRaises(zipfile.LargeZipFile): + zipf.writestr("foo%08d" % numfiles, b'') + self.assertEqual(len(zipf.namelist()), numfiles) + zipf.close() + + zipf = zipfile2.ZipFile(TESTFN, "a", self.compression, + allowZip64=False) + zipf.debug = 100 + self.assertEqual(len(zipf.namelist()), numfiles) + with self.assertRaises(zipfile.LargeZipFile): + zipf.writestr("foo%08d" % numfiles, b'') + self.assertEqual(len(zipf.namelist()), numfiles) + zipf.close() + + zipf = zipfile2.ZipFile(TESTFN, "a", self.compression, + allowZip64=True) + zipf.debug = 100 + self.assertEqual(len(zipf.namelist()), numfiles) + numfiles2 = 15 + for i in range(numfiles, numfiles2): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles2) + zipf.close() + + zipf2 = zipfile2.ZipFile(TESTFN, "r", self.compression) + self.assertEqual(len(zipf2.namelist()), numfiles2) + for i in range(numfiles2): + content = zipf2.read("foo%08d" % i).decode('ascii') + self.assertEqual(content, "%d" % (i**3 % 57)) + zipf2.close() + + def tearDown(self): + zipfile.ZIP64_LIMIT = self._limit + zipfile.ZIP_FILECOUNT_LIMIT = self._filecount_limit + unlink(TESTFN) + unlink(TESTFN2) + + +class StoredTestZip64InSmallFiles(AbstractTestZip64InSmallFiles, + unittest.TestCase): + compression = zipfile.ZIP_STORED + + def large_file_exception_test(self, f, compression): + with zipfile2.ZipFile(f, "w", compression, allowZip64=False) as zipfp: + self.assertRaises(zipfile.LargeZipFile, + zipfp.write, TESTFN, "another.name") + + def large_file_exception_test2(self, f, compression): + with zipfile2.ZipFile(f, "w", compression, allowZip64=False) as zipfp: + self.assertRaises(zipfile.LargeZipFile, + zipfp.writestr, "another.name", self.data) + + def test_large_file_exception(self): + for f in get_files(self): + self.large_file_exception_test(f, zipfile.ZIP_STORED) + self.large_file_exception_test2(f, zipfile.ZIP_STORED) + + def test_absolute_arcnames(self): + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED, + allowZip64=True) as zipfp: + zipfp.write(TESTFN, "/absolute") + + with zipfile2.ZipFile(TESTFN2, "r", zipfile.ZIP_STORED) as zipfp: + self.assertEqual(zipfp.namelist(), ["absolute"]) + + def test_append(self): + # Test that appending to the Zip64 archive doesn't change + # extra fields of existing entries. + with zipfile2.ZipFile(TESTFN2, "w", allowZip64=True) as zipfp: + zipfp.writestr("strfile", self.data) + with zipfile2.ZipFile(TESTFN2, "r", allowZip64=True) as zipfp: + zinfo = zipfp.getinfo("strfile") + extra = zinfo.extra + with zipfile2.ZipFile(TESTFN2, "a", allowZip64=True) as zipfp: + zipfp.writestr("strfile2", self.data) + with zipfile2.ZipFile(TESTFN2, "r", allowZip64=True) as zipfp: + zinfo = zipfp.getinfo("strfile") + self.assertEqual(zinfo.extra, extra) + + def make_zip64_file( + self, file_size_64_set=False, file_size_extra=False, + compress_size_64_set=False, compress_size_extra=False, + header_offset_64_set=False, header_offset_extra=False, + ): + """Generate bytes sequence for a zip with (incomplete) zip64 data. + + The actual values (not the zip 64 0xffffffff values) stored in the file + are: + file_size: 8 + compress_size: 8 + header_offset: 0 + """ + actual_size = 8 + actual_header_offset = 0 + local_zip64_fields = [] + central_zip64_fields = [] + + file_size = actual_size + if file_size_64_set: + file_size = 0xffffffff + if file_size_extra: + local_zip64_fields.append(actual_size) + central_zip64_fields.append(actual_size) + file_size = struct.pack("e|f"g?h*i', ','), r'a\b,c_d_e_f_g_h_i') + self.assertEqual(san('../../foo../../ba..r', '/'), r'foo/ba..r') + + def test_extract_hackers_arcnames_common_cases(self): + common_hacknames = [ + ('../foo/bar', 'foo/bar'), + ('foo/../bar', 'foo/bar'), + ('foo/../../bar', 'foo/bar'), + ('foo/bar/..', 'foo/bar'), + ('./../foo/bar', 'foo/bar'), + ('/foo/bar', 'foo/bar'), + ('/foo/../bar', 'foo/bar'), + ('/foo/../../bar', 'foo/bar'), + ] + self._test_extract_hackers_arcnames(common_hacknames) + + @unittest.skipIf(os.path.sep != '\\', 'Requires \\ as path separator.') + def test_extract_hackers_arcnames_windows_only(self): + """Test combination of path fixing and windows name sanitization.""" + windows_hacknames = [ + (r'..\foo\bar', 'foo/bar'), + (r'..\/foo\/bar', 'foo/bar'), + (r'foo/\..\/bar', 'foo/bar'), + (r'foo\/../\bar', 'foo/bar'), + (r'C:foo/bar', 'foo/bar'), + (r'C:/foo/bar', 'foo/bar'), + (r'C://foo/bar', 'foo/bar'), + (r'C:\foo\bar', 'foo/bar'), + (r'//conky/mountpoint/foo/bar', 'foo/bar'), + (r'\\conky\mountpoint\foo\bar', 'foo/bar'), + (r'///conky/mountpoint/foo/bar', 'conky/mountpoint/foo/bar'), + (r'\\\conky\mountpoint\foo\bar', 'conky/mountpoint/foo/bar'), + (r'//conky//mountpoint/foo/bar', 'conky/mountpoint/foo/bar'), + (r'\\conky\\mountpoint\foo\bar', 'conky/mountpoint/foo/bar'), + (r'//?/C:/foo/bar', 'foo/bar'), + (r'\\?\C:\foo\bar', 'foo/bar'), + (r'C:/../C:/foo/bar', 'C_/foo/bar'), + (r'a:b\ce|f"g?h*i', 'b/c_d_e_f_g_h_i'), + ('../../foo../../ba..r', 'foo/ba..r'), + ] + self._test_extract_hackers_arcnames(windows_hacknames) + + @unittest.skipIf(os.path.sep != '/', r'Requires / as path separator.') + def test_extract_hackers_arcnames_posix_only(self): + posix_hacknames = [ + ('//foo/bar', 'foo/bar'), + ('../../foo../../ba..r', 'foo../ba..r'), + (r'foo/..\bar', r'foo/..\bar'), + ] + self._test_extract_hackers_arcnames(posix_hacknames) + + def _test_extract_hackers_arcnames(self, hacknames): + for arcname, fixedname in hacknames: + content = b'foobar' + arcname.encode() + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipfp: + zinfo = zipfile.ZipInfo() + # preserve backslashes + zinfo.filename = arcname + zinfo.external_attr = 0o600 << 16 + zipfp.writestr(zinfo, content) + + arcname = arcname.replace(os.sep, "/") + targetpath = os.path.join('target', 'subdir', 'subsub') + correctfile = os.path.join(targetpath, *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname, targetpath) + self.assertEqual(writtenfile, correctfile, + msg='extract %r: %r != %r' % + (arcname, writtenfile, correctfile)) + self.check_file(correctfile, content) + rmtree('target') + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall(targetpath) + self.check_file(correctfile, content) + rmtree('target') + + correctfile = os.path.join(os.getcwd(), *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname) + self.assertEqual(writtenfile, correctfile, + msg="extract %r" % arcname) + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall() + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + unlink(TESTFN2) + + +class OtherTests(unittest.TestCase): + def test_open_via_zip_info(self): + # Create the ZIP archive + with zipfile2.ZipFile( + TESTFN2, "w", zipfile.ZIP_STORED, low_level=True) as zipfp: + zipfp.writestr("name", "foo") + with self.assertWarns(UserWarning): + zipfp.writestr("name", "bar") + self.assertEqual(zipfp.namelist(), ["name"] * 2) + + with zipfile2.ZipFile(TESTFN2, "r", low_level=True) as zipfp: + infos = zipfp.infolist() + data = b"" + for info in infos: + with zipfp.open(info) as zipopen: + data += zipopen.read() + self.assertIn(data, {b"foobar", b"barfoo"}) + data = b"" + for info in infos: + data += zipfp.read(info) + self.assertIn(data, {b"foobar", b"barfoo"}) + + def test_writestr_extended_local_header_issue1202(self): + with zipfile2.ZipFile(TESTFN2, 'w') as orig_zip: + for data in 'abcdefghijklmnop': + zinfo = zipfile.ZipInfo(data) + zinfo.flag_bits |= 0x08 # Include an extended local header. + orig_zip.writestr(zinfo, data) + + def test_close(self): + """Check that the zipfile is closed after the 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + def test_close_on_exception(self): + """Check that the zipfile is closed if an exception is raised in the + 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + + try: + with zipfile2.ZipFile(TESTFN2, "r") as zipfp2: + raise zipfile.BadZipFile() + except zipfile.BadZipFile: + self.assertIsNone(zipfp2.fp, 'zipfp is not closed') + + def test_unsupported_version(self): + # File has an extract_version of 120 + data = (b'PK\x03\x04x\x00\x00\x00\x00\x00!p\xa1@\x00\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x01\x00\x00\x00xPK\x01\x02x\x03x\x00\x00\x00\x00' + b'\x00!p\xa1@\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x01\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00\x00xPK\x05\x06' + b'\x00\x00\x00\x00\x01\x00\x01\x00/\x00\x00\x00\x1f\x00\x00\x00\x00\x00') + + self.assertRaises(NotImplementedError, zipfile2.ZipFile, + io.BytesIO(data), 'r') + + @requires_zlib() + def test_read_unicode_filenames(self): + # bug #10801 + fname = findfile('zip_cp437_header.zip') + with zipfile2.ZipFile(fname) as zipfp: + for name in zipfp.namelist(): + zipfp.open(name).close() + + def test_write_unicode_filenames(self): + with zipfile2.ZipFile(TESTFN, "w") as zf: + zf.writestr("foo.txt", "Test for unicode filename") + zf.writestr("\xf6.txt", "Test for unicode filename") + self.assertIsInstance(zf.infolist()[0].filename, str) + + with zipfile2.ZipFile(TESTFN, "r") as zf: + self.assertEqual(zf.filelist[0].filename, "foo.txt") + self.assertEqual(zf.filelist[1].filename, "\xf6.txt") + + def test_read_after_write_unicode_filenames(self): + with zipfile2.ZipFile(TESTFN2, 'w') as zipfp: + zipfp.writestr('приклад', b'sample') + self.assertEqual(zipfp.read('приклад'), b'sample') + + def test_exclusive_create_zip_file(self): + """Test exclusive creating a new zipfile.""" + unlink(TESTFN2) + filename = 'testfile.txt' + content = b'hello, world. this is some content.' + with zipfile2.ZipFile(TESTFN2, "x", zipfile.ZIP_STORED) as zipfp: + zipfp.writestr(filename, content) + with self.assertRaises(FileExistsError): + zipfile2.ZipFile(TESTFN2, "x", zipfile.ZIP_STORED) + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertEqual(zipfp.namelist(), [filename]) + self.assertEqual(zipfp.read(filename), content) + + def test_create_non_existent_file_for_append(self): + if os.path.exists(TESTFN): + os.unlink(TESTFN) + + filename = 'testfile.txt' + content = b'hello, world. this is some content.' + + try: + with zipfile2.ZipFile(TESTFN, 'a') as zf: + zf.writestr(filename, content) + except OSError: + self.fail('Could not append data to a non-existent zip file.') + + self.assertTrue(os.path.exists(TESTFN)) + + with zipfile2.ZipFile(TESTFN, 'r') as zf: + self.assertEqual(zf.read(filename), content) + + def test_close_erroneous_file(self): + # This test checks that the ZipFile constructor closes the file object + # it opens if there's an error in the file. If it doesn't, the + # traceback holds a reference to the ZipFile object and, indirectly, + # the file object. + # On Windows, this causes the os.unlink() call to fail because the + # underlying file is still open. This is SF bug #412214. + # + with open(TESTFN, "w", encoding="utf-8") as fp: + fp.write("this is not a legal zip file\n") + try: + zf = zipfile2.ZipFile(TESTFN) + except zipfile.BadZipFile: + pass + + def test_is_zip_erroneous_file(self): + """Check that is_zipfile() correctly identifies non-zip files.""" + # - passing a filename + with open(TESTFN, "w", encoding='utf-8') as fp: + fp.write("this is not a legal zip file\n") + self.assertFalse(zipfile.is_zipfile(TESTFN)) + # - passing a path-like object + self.assertFalse(zipfile.is_zipfile(pathlib.Path(TESTFN))) + # - passing a file object + with open(TESTFN, "rb") as fp: + self.assertFalse(zipfile.is_zipfile(fp)) + # - passing a file-like object + fp = io.BytesIO() + fp.write(b"this is not a legal zip file\n") + self.assertFalse(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + self.assertFalse(zipfile.is_zipfile(fp)) + + def test_damaged_zipfile(self): + """Check that zipfiles with missing bytes at the end raise BadZipFile.""" + # - Create a valid zip file + fp = io.BytesIO() + with zipfile2.ZipFile(fp, mode="w") as zipf: + zipf.writestr("foo.txt", b"O, for a Muse of Fire!") + zipfiledata = fp.getvalue() + + # - Now create copies of it missing the last N bytes and make sure + # a BadZipFile exception is raised when we try to open it + for N in range(len(zipfiledata)): + fp = io.BytesIO(zipfiledata[:N]) + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, fp) + + def test_is_zip_valid_file(self): + """Check that is_zipfile() correctly identifies zip files.""" + # - passing a filename + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt", b"O, for a Muse of Fire!") + + self.assertTrue(zipfile.is_zipfile(TESTFN)) + # - passing a file object + with open(TESTFN, "rb") as fp: + self.assertTrue(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + zip_contents = fp.read() + # - passing a file-like object + fp = io.BytesIO() + fp.write(zip_contents) + self.assertTrue(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + self.assertTrue(zipfile.is_zipfile(fp)) + + def test_non_existent_file_raises_OSError(self): + # make sure we don't raise an AttributeError when a partially-constructed + # ZipFile instance is finalized; this tests for regression on SF tracker + # bug #403871. + + # The bug we're testing for caused an AttributeError to be raised + # when a ZipFile instance was created for a file that did not + # exist; the .fp member was not initialized but was needed by the + # __del__() method. Since the AttributeError is in the __del__(), + # it is ignored, but the user should be sufficiently annoyed by + # the message on the output that regression will be noticed + # quickly. + self.assertRaises(OSError, zipfile2.ZipFile, TESTFN) + + def test_empty_file_raises_BadZipFile(self): + f = open(TESTFN, 'w', encoding='utf-8') + f.close() + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN) + + with open(TESTFN, 'w', encoding='utf-8') as fp: + fp.write("short file") + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN) + + def test_negative_central_directory_offset_raises_BadZipFile(self): + # Zip file containing an empty EOCD record + buffer = bytearray(b'PK\x05\x06' + b'\0'*18) + + # Set the size of the central directory bytes to become 1, + # causing the central directory offset to become negative + for dirsize in 1, 2**32-1: + buffer[12:16] = struct.pack(' os.path.getsize(TESTFN)) + with zipfile2.ZipFile(TESTFN,mode="r") as zipf: + self.assertEqual(zipf.comment, b"shorter comment") + + def test_unicode_comment(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with self.assertRaises(TypeError): + zipf.comment = "this is an error" + + def test_change_comment_in_empty_archive(self): + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertFalse(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_change_comment_in_nonempty_archive(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertTrue(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_empty_zipfile(self): + # Check that creating a file in 'w' or 'a' mode and closing without + # adding any files to the archives creates a valid empty ZIP file + zipf = zipfile2.ZipFile(TESTFN, mode="w") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except zipfile.BadZipFile: + self.fail("Unable to create empty ZIP file in 'w' mode") + + zipf = zipfile2.ZipFile(TESTFN, mode="a") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except: + self.fail("Unable to create empty ZIP file in 'a' mode") + + def test_open_empty_file(self): + # Issue 1710703: Check that opening a file with less than 22 bytes + # raises a BadZipFile exception (rather than the previously unhelpful + # OSError) + f = open(TESTFN, 'w', encoding='utf-8') + f.close() + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN, 'r') + + def test_create_zipinfo_before_1980(self): + self.assertRaises(ValueError, + zipfile.ZipInfo, 'seventies', (1979, 1, 1, 0, 0, 0)) + + def test_create_empty_zipinfo_repr(self): + """Before bpo-26185, repr() on empty ZipInfo object was failing.""" + zi = zipfile.ZipInfo(filename="empty") + self.assertEqual(repr(zi), "") + + def test_create_empty_zipinfo_default_attributes(self): + """Ensure all required attributes are set.""" + zi = zipfile.ZipInfo() + self.assertEqual(zi.orig_filename, "NoName") + self.assertEqual(zi.filename, "NoName") + self.assertEqual(zi.date_time, (1980, 1, 1, 0, 0, 0)) + self.assertEqual(zi.compress_type, zipfile.ZIP_STORED) + self.assertEqual(zi.comment, b"") + self.assertEqual(zi.extra, b"") + self.assertIn(zi.create_system, (0, 3)) + self.assertEqual(zi.create_version, zipfile.DEFAULT_VERSION) + self.assertEqual(zi.extract_version, zipfile.DEFAULT_VERSION) + self.assertEqual(zi.reserved, 0) + self.assertEqual(zi.flag_bits, 0) + self.assertEqual(zi.volume, 0) + self.assertEqual(zi.internal_attr, 0) + self.assertEqual(zi.external_attr, 0) + + # Before bpo-26185, both were missing + self.assertEqual(zi.file_size, 0) + self.assertEqual(zi.compress_size, 0) + + def test_zipfile_with_short_extra_field(self): + """If an extra field in the header is less than 4 bytes, skip it.""" + zipdata = ( + b'PK\x03\x04\x14\x00\x00\x00\x00\x00\x93\x9b\xad@\x8b\x9e' + b'\xd9\xd3\x01\x00\x00\x00\x01\x00\x00\x00\x03\x00\x03\x00ab' + b'c\x00\x00\x00APK\x01\x02\x14\x03\x14\x00\x00\x00\x00' + b'\x00\x93\x9b\xad@\x8b\x9e\xd9\xd3\x01\x00\x00\x00\x01\x00\x00' + b'\x00\x03\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\xa4\x81\x00' + b'\x00\x00\x00abc\x00\x00PK\x05\x06\x00\x00\x00\x00' + b'\x01\x00\x01\x003\x00\x00\x00%\x00\x00\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(zipdata), 'r') as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertIsNone(zipf.testzip()) + + def test_open_conflicting_handles(self): + # It's only possible to open one writable file handle at a time + msg1 = b"It's fun to charter an accountant!" + msg2 = b"And sail the wide accountant sea" + msg3 = b"To find, explore the funds offshore" + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipf: + with zipf.open('foo', mode='w') as w2: + w2.write(msg1) + with zipf.open('bar', mode='w') as w1: + with self.assertRaises(ValueError): + zipf.open('handle', mode='w') + with self.assertRaises(ValueError): + zipf.open('foo', mode='r') + with self.assertRaises(ValueError): + zipf.writestr('str', 'abcde') + with self.assertRaises(ValueError): + zipf.write(__file__, 'file') + with self.assertRaises(ValueError): + zipf.close() + w1.write(msg2) + with zipf.open('baz', mode='w') as w2: + w2.write(msg3) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipf: + self.assertEqual(zipf.read('foo'), msg1) + self.assertEqual(zipf.read('bar'), msg2) + self.assertEqual(zipf.read('baz'), msg3) + self.assertEqual(zipf.namelist(), ['foo', 'bar', 'baz']) + + def test_seek_tell(self): + # Test seek functionality + txt = b"Where's Bruce?" + bloc = txt.find(b"Bruce") + # Check seek on a file + with zipfile2.ZipFile(TESTFN, "w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(TESTFN, "r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + # Check seek on memory file + data = io.BytesIO() + with zipfile2.ZipFile(data, mode="w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(data, mode="r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + + @requires_bz2() + def test_decompress_without_3rd_party_library(self): + data = b'PK\x05\x06\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + zip_file = io.BytesIO(data) + with zipfile2.ZipFile(zip_file, 'w', compression=zipfile.ZIP_BZIP2) as zf: + zf.writestr('a.txt', b'a') + with mock.patch('zipfile.bz2', None): + with zipfile2.ZipFile(zip_file) as zf: + self.assertRaises(RuntimeError, zf.extract, 'a.txt') + + @requires_zlib() + def test_full_overlap(self): + data = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e' + b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00a\xed' + b'\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\d\x0b`P' + b'K\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2' + b'\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00aPK' + b'\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e' + b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00bPK\x05' + b'\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00\x00/\x00\x00' + b'\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertEqual(zipf.namelist(), ['a', 'b']) + zi = zipf.getinfo('a') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + zi = zipf.getinfo('b') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + self.assertEqual(len(zipf.read('a')), 1033) + with self.assertRaisesRegex(zipfile.BadZipFile, 'File name.*differ'): + zipf.read('b') + + @requires_zlib() + @unittest.skipIf(sys.version_info < (3, 10, 14), "Behaviour not supported in Python < 3.10.14") + def test_quoted_overlap(self): + data = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05Y\xfc' + b'8\x044\x00\x00\x00(\x04\x00\x00\x01\x00\x00\x00a\x00' + b'\x1f\x00\xe0\xffPK\x03\x04\x14\x00\x00\x00\x08\x00\xa0l' + b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00' + b'\x00\x00b\xed\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\' + b'd\x0b`PK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0' + b'lH\x05Y\xfc8\x044\x00\x00\x00(\x04\x00\x00\x01' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x00aPK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0l' + b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00$\x00\x00\x00' + b'bPK\x05\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00' + b'\x00S\x00\x00\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertEqual(zipf.namelist(), ['a', 'b']) + zi = zipf.getinfo('a') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 52) + self.assertEqual(zi.file_size, 1064) + zi = zipf.getinfo('b') + self.assertEqual(zi.header_offset, 36) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + with self.assertRaisesRegex(zipfile.BadZipFile, 'Overlapped entries'): + zipf.read('a') + self.assertEqual(len(zipf.read('b')), 1033) + + def tearDown(self): + unlink(TESTFN) + unlink(TESTFN2) + + +class AbstractBadCrcTests: + def test_testzip_with_bad_crc(self): + """Tests that files with bad CRCs return their name from testzip.""" + zipdata = self.zip_with_bad_crc + + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertEqual('afile', zipf.testzip()) + + def test_read_with_bad_crc(self): + """Tests that files with bad CRCs raise a BadZipFile exception when read.""" + zipdata = self.zip_with_bad_crc + + # Using ZipFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + self.assertRaises(zipfile.BadZipFile, zipf.read, 'afile') + + # Using ZipExtFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + self.assertRaises(zipfile.BadZipFile, corrupt_file.read) + + # Same with small reads (in order to exercise the buffering logic) + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + corrupt_file.MIN_READ_SIZE = 2 + with self.assertRaises(zipfile.BadZipFile): + while corrupt_file.read(2): + pass + + +class StoredBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_STORED + zip_with_bad_crc = ( + b'PK\003\004\024\0\0\0\0\0 \213\212;:r' + b'\253\377\f\0\0\0\f\0\0\0\005\0\0\000af' + b'ilehello,AworldP' + b'K\001\002\024\003\024\0\0\0\0\0 \213\212;:' + b'r\253\377\f\0\0\0\f\0\0\0\005\0\0\0\0' + b'\0\0\0\0\0\0\0\200\001\0\0\0\000afi' + b'lePK\005\006\0\0\0\0\001\0\001\0003\000' + b'\0\0/\0\0\0\0\0') + +@requires_zlib() +class DeflateBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_DEFLATED + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00n}\x0c=FA' + b'KE\x10\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\xcbH\xcd\xc9\xc9W(\xcf/\xcaI\xc9\xa0' + b'=\x13\x00PK\x01\x02\x14\x03\x14\x00\x00\x00\x08\x00n' + b'}\x0c=FAKE\x10\x00\x00\x00n\x00\x00\x00\x05' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00' + b'\x00afilePK\x05\x06\x00\x00\x00\x00\x01\x00' + b'\x01\x003\x00\x00\x003\x00\x00\x00\x00\x00') + +@requires_bz2() +class Bzip2BadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_BZIP2 + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0c\x00nu\x0c=FA' + b'KE8\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ileBZh91AY&SY\xd4\xa8\xca' + b'\x7f\x00\x00\x0f\x11\x80@\x00\x06D\x90\x80 \x00 \xa5' + b'P\xd9!\x03\x03\x13\x13\x13\x89\xa9\xa9\xc2u5:\x9f' + b'\x8b\xb9"\x9c(HjTe?\x80PK\x01\x02\x14' + b'\x03\x14\x03\x00\x00\x0c\x00nu\x0c=FAKE8' + b'\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x00 \x80\x80\x81\x00\x00\x00\x00afilePK' + b'\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00\x00[\x00' + b'\x00\x00\x00\x00') + +@requires_lzma() +class LzmaBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_LZMA + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\t\x04\x05\x00]\x00\x00\x00\x04\x004\x19I' + b'\xee\x8d\xe9\x17\x89:3`\tq!.8\x00PK' + b'\x01\x02\x14\x03\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00 \x80\x80\x81\x00\x00\x00\x00afil' + b'ePK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00' + b'\x00>\x00\x00\x00\x00\x00') + + +class DecryptionTests(unittest.TestCase): + """Check that ZIP decryption works. Since the library does not + support encryption at the moment, we use a pre-generated encrypted + ZIP file.""" + + data = ( + b'PK\x03\x04\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00\x1a\x00' + b'\x00\x00\x08\x00\x00\x00test.txt\xfa\x10\xa0gly|\xfa-\xc5\xc0=\xf9y' + b'\x18\xe0\xa8r\xb3Z}Lg\xbc\xae\xf9|\x9b\x19\xe4\x8b\xba\xbb)\x8c\xb0\xdbl' + b'PK\x01\x02\x14\x00\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00' + b'\x1a\x00\x00\x00\x08\x00\x00\x00\x00\x00\x00\x00\x01\x00 \x00\xb6\x81' + b'\x00\x00\x00\x00test.txtPK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x006\x00' + b'\x00\x00L\x00\x00\x00\x00\x00' ) + data2 = ( + b'PK\x03\x04\x14\x00\t\x00\x08\x00\xcf}38xu\xaa\xb2\x14\x00\x00\x00\x00\x02' + b'\x00\x00\x04\x00\x15\x00zeroUT\t\x00\x03\xd6\x8b\x92G\xda\x8b\x92GUx\x04' + b'\x00\xe8\x03\xe8\x03\xc7:8: EncodingWarning:") + self.assertRegex(warnings[1], rb"^:9: EncodingWarning:") + + def test_open_write(self): + """ + If the zipfile is open for write, it should be possible to + write bytes or text to it. + """ + zf = zipfile.Path(zipfile2.ZipFile(io.BytesIO(), mode='w')) + with zf.joinpath('file.bin').open('wb') as strm: + strm.write(b'binary contents') + with zf.joinpath('file.txt').open('w', encoding="utf-8") as strm: + strm.write('text file') + + def test_open_extant_directory(self): + """ + Attempting to open a directory raises IsADirectoryError. + """ + zf = zipfile.Path(add_dirs(build_alpharep_fixture())) + with self.assertRaises(IsADirectoryError): + zf.joinpath('b').open() + + @pass_alpharep + def test_open_binary_invalid_args(self, alpharep): + root = zipfile.Path(alpharep) + with self.assertRaises(ValueError): + root.joinpath('a.txt').open('rb', encoding='utf-8') + with self.assertRaises(ValueError): + root.joinpath('a.txt').open('rb', 'utf-8') + + def test_open_missing_directory(self): + """ + Attempting to open a missing directory raises FileNotFoundError. + """ + zf = zipfile.Path(add_dirs(build_alpharep_fixture())) + with self.assertRaises(FileNotFoundError): + zf.joinpath('z').open() + + @pass_alpharep + def test_read(self, alpharep): + root = zipfile.Path(alpharep) + a, b, g = root.iterdir() + assert a.read_text(encoding="utf-8") == "content of a" + a.read_text("utf-8") # No positional arg TypeError per gh-101144. + assert a.read_bytes() == b"content of a" + + @pass_alpharep + def test_joinpath(self, alpharep): + root = zipfile.Path(alpharep) + a = root.joinpath("a.txt") + assert a.is_file() + e = root.joinpath("b").joinpath("d").joinpath("e.txt") + assert e.read_text(encoding="utf-8") == "content of e" + + @pass_alpharep + def test_joinpath_multiple(self, alpharep): + root = zipfile.Path(alpharep) + e = root.joinpath("b", "d", "e.txt") + assert e.read_text(encoding="utf-8") == "content of e" + + @pass_alpharep + def test_traverse_truediv(self, alpharep): + root = zipfile.Path(alpharep) + a = root / "a.txt" + assert a.is_file() + e = root / "b" / "d" / "e.txt" + assert e.read_text(encoding="utf-8") == "content of e" + + @pass_alpharep + def test_traverse_simplediv(self, alpharep): + """ + Disable the __future__.division when testing traversal. + """ + code = compile( + source="zipfile.Path(alpharep) / 'a'", + filename="(test)", + mode="eval", + dont_inherit=True, + ) + eval(code) + + @pass_alpharep + def test_pathlike_construction(self, alpharep): + """ + zipfile.Path should be constructable from a path-like object + """ + zipfile_ondisk = self.zipfile_ondisk(alpharep) + pathlike = pathlib.Path(str(zipfile_ondisk)) + zipfile.Path(pathlike) + + @pass_alpharep + def test_traverse_pathlike(self, alpharep): + root = zipfile.Path(alpharep) + root / pathlib.Path("a") + + @pass_alpharep + def test_parent(self, alpharep): + root = zipfile.Path(alpharep) + assert (root / 'a').parent.at == '' + assert (root / 'a' / 'b').parent.at == 'a/' + + @pass_alpharep + def test_dir_parent(self, alpharep): + root = zipfile.Path(alpharep) + assert (root / 'b').parent.at == '' + assert (root / 'b/').parent.at == '' + + @pass_alpharep + def test_missing_dir_parent(self, alpharep): + root = zipfile.Path(alpharep) + assert (root / 'missing dir/').parent.at == '' + + @pass_alpharep + def test_mutability(self, alpharep): + """ + If the underlying zipfile is changed, the Path object should + reflect that change. + """ + root = zipfile.Path(alpharep) + a, b, g = root.iterdir() + alpharep.writestr('foo.txt', 'foo') + alpharep.writestr('bar/baz.txt', 'baz') + assert any(child.name == 'foo.txt' for child in root.iterdir()) + assert (root / 'foo.txt').read_text(encoding="utf-8") == 'foo' + (baz,) = (root / 'bar').iterdir() + assert baz.read_text(encoding="utf-8") == 'baz' + + HUGE_ZIPFILE_NUM_ENTRIES = 2 ** 13 + + def huge_zipfile(self): + """Create a read-only zipfile with a huge number of entries entries.""" + strm = io.BytesIO() + zf = zipfile2.ZipFile(strm, "w") + for entry in map(str, range(self.HUGE_ZIPFILE_NUM_ENTRIES)): + zf.writestr(entry, entry) + zf.mode = 'r' + return zf + + def test_joinpath_constant_time(self): + """ + Ensure joinpath on items in zipfile is linear time. + """ + root = zipfile.Path(self.huge_zipfile()) + entries = jaraco.itertools.Counter(root.iterdir()) + for entry in entries: + entry.joinpath('suffix') + # Check the file iterated all items + assert entries.count == self.HUGE_ZIPFILE_NUM_ENTRIES + + # @func_timeout.func_set_timeout(3) + def test_implied_dirs_performance(self): + data = ['/'.join(string.ascii_lowercase + str(n)) for n in range(10000)] + zipfile.CompleteDirs._implied_dirs(data) + + @pass_alpharep + def test_read_does_not_close(self, alpharep): + alpharep = self.zipfile_ondisk(alpharep) + with zipfile2.ZipFile(alpharep) as file: + for rep in range(2): + zipfile.Path(file, 'a.txt').read_text(encoding="utf-8") + + @pass_alpharep + def test_subclass(self, alpharep): + class Subclass(zipfile.Path): + pass + + root = Subclass(alpharep) + assert isinstance(root / 'b', Subclass) + + @pass_alpharep + def test_filename(self, alpharep): + root = zipfile.Path(alpharep) + assert root.filename == pathlib.Path('alpharep.zip') + + @pass_alpharep + def test_root_name(self, alpharep): + """ + The name of the root should be the name of the zipfile + """ + root = zipfile.Path(alpharep) + assert root.name == 'alpharep.zip' == root.filename.name + + @pass_alpharep + def test_root_parent(self, alpharep): + root = zipfile.Path(alpharep) + assert root.parent == pathlib.Path('.') + root.root.filename = 'foo/bar.zip' + assert root.parent == pathlib.Path('foo') + + @pass_alpharep + def test_root_unnamed(self, alpharep): + """ + It is an error to attempt to get the name + or parent of an unnamed zipfile. + """ + alpharep.filename = None + root = zipfile.Path(alpharep) + with self.assertRaises(TypeError): + root.name + with self.assertRaises(TypeError): + root.parent + + # .name and .parent should still work on subs + sub = root / "b" + assert sub.name == "b" + assert sub.parent + + @pass_alpharep + def test_inheritance(self, alpharep): + cls = type('PathChild', (zipfile.Path,), {}) + for alpharep in self.zipfile_alpharep(): + file = cls(alpharep).joinpath('some dir').parent + assert isinstance(file, cls) + + @pass_alpharep + def test_extract_orig_with_implied_dirs(self, alpharep): + """ + A zip file wrapped in a Path should extract even with implied dirs. + """ + source_path = self.zipfile_ondisk(alpharep) + zf = zipfile2.ZipFile(source_path) + # wrap the zipfile for its side effect + zipfile.Path(zf) + zf.extractall(source_path.parent) + + @unittest.skipIf(sys.version_info < (3, 10, 15), "Issue fixed in Python >= 3.10.15") + def test_malformed_paths(self): + """ + Path should handle malformed paths gracefully. + + Paths with leading slashes are not visible. + + Paths with dots are treated like regular files. + """ + data = io.BytesIO() + zf = zipfile2.ZipFile(data, "w") + zf.writestr("/one-slash.txt", b"content") + zf.writestr("//two-slash.txt", b"content") + zf.writestr("../parent.txt", b"content") + zf.filename = '' + root = zipfile.Path(zf) + assert list(map(str, root.iterdir())) == ['../'] + assert root.joinpath('..').joinpath('parent.txt').read_bytes() == b'content' + + @unittest.skipIf(sys.version_info < (3, 10, 15), "Issue fixed in Python >= 3.10.15") + def test_unsupported_names(self): + """ + Path segments with special characters are readable. + + On some platforms or file systems, characters like + ``:`` and ``?`` are not allowed, but they are valid + in the zip file. + """ + data = io.BytesIO() + zf = zipfile2.ZipFile(data, "w") + zf.writestr("path?", b"content") + zf.writestr("V: NMS.flac", b"fLaC...") + zf.filename = '' + root = zipfile.Path(zf) + contents = root.iterdir() + assert next(contents).name == 'path?' + item = next(contents) + assert item.name == 'V: NMS.flac', item.name + assert root.joinpath('V: NMS.flac').read_bytes() == b"fLaC..." + + @unittest.skipIf(sys.version_info < (3, 10, 15), "Issue fixed in Python >= 3.10.15") + def test_backslash_not_separator(self): + """ + In a zip file, backslashes are not separators. + """ + data = io.BytesIO() + zf = zipfile2.ZipFile(data, "w") + zf.writestr(DirtyZipInfo.for_name("foo\\bar", zf), b"content") + zf.filename = '' + root = zipfile.Path(zf) + (first,) = root.iterdir() + assert not first.is_dir() + assert first.name == 'foo\\bar', first.name + + +class DirtyZipInfo(zipfile.ZipInfo): + """ + Bypass name sanitization. + """ + + def __init__(self, filename, *args, **kwargs): + super().__init__(filename, *args, **kwargs) + self.filename = filename + + @classmethod + def for_name(cls, name, archive): + """ + Construct the same way that ZipFile.writestr does. + + TODO: extract this functionality and re-use + """ + self = cls(filename=name, date_time=time.localtime(time.time())[:6]) + self.compress_type = archive.compression + self.compress_level = archive.compresslevel + if self.filename.endswith('/'): # pragma: no cover + self.external_attr = 0o40775 << 16 # drwxrwxr-x + self.external_attr |= 0x10 # MS-DOS directory flag + else: + self.external_attr = 0o600 << 16 # ?rw------- + return self + + +class StripExtraTests(unittest.TestCase): + # Note: all of the "z" characters are technically invalid, but up + # to 3 bytes at the end of the extra will be passed through as they + # are too short to encode a valid extra. + + ZIP64_EXTRA = 1 + + def test_no_data(self): + s = struct.Struct("", "exec") +raise_src = 'def do_raise(): raise TypeError\n' + +def make_pyc(co, mtime, size): + data = marshal.dumps(co) + pyc = (importlib.util.MAGIC_NUMBER + + struct.pack("", "exec"), NOW, len(src)) + files = {TESTMOD + pyc_ext: (NOW, pyc), + "some.data": (NOW, "some data")} + self.doTest(pyc_ext, files, TESTMOD) + + def testDefaultOptimizationLevel(self): + # zipimport should use the default optimization level (#28131) + src = """if 1: # indent hack + def test(val): + assert(val) + return val\n""" + files = {TESTMOD + '.py': (NOW, src)} + self.makeZip(files) + sys.path.insert(0, TEMP_ZIP) + mod = importlib.import_module(TESTMOD) + self.assertEqual(mod.test(1), 1) + if __debug__: + self.assertRaises(AssertionError, mod.test, False) + else: + self.assertEqual(mod.test(0), 0) + + def testImport_WithStuff(self): + # try importing from a zipfile which contains additional + # stuff at the beginning of the file + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, + stuff=b"Some Stuff"*31) + + def assertModuleSource(self, module): + self.assertEqual(inspect.getsource(module), test_src) + + def testGetSource(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, call=self.assertModuleSource) + + def testGetCompiledSource(self): + pyc = make_pyc(compile(test_src, "", "exec"), NOW, len(test_src)) + files = {TESTMOD + ".py": (NOW, test_src), + TESTMOD + pyc_ext: (NOW, pyc)} + self.doTest(pyc_ext, files, TESTMOD, call=self.assertModuleSource) + + def runDoctest(self, callback): + files = {TESTMOD + ".py": (NOW, test_src), + "xyz.txt": (NOW, ">>> log.append(True)\n")} + self.doTest(".py", files, TESTMOD, call=callback) + + def doDoctestFile(self, module): + log = [] + old_master, doctest.master = doctest.master, None + try: + doctest.testfile( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ) + finally: + doctest.master = old_master + self.assertEqual(log,[True]) + + def testDoctestFile(self): + self.runDoctest(self.doDoctestFile) + + def doDoctestSuite(self, module): + log = [] + doctest.DocFileTest( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ).run() + self.assertEqual(log,[True]) + + def testDoctestSuite(self): + self.runDoctest(self.doDoctestSuite) + + def doTraceback(self, module): + try: + module.do_raise() + except Exception as e: + tb = e.__traceback__.tb_next + + f,lno,n,line = extract_tb(tb, 1)[0] + self.assertEqual(line, raise_src.strip()) + + f,lno,n,line = extract_stack(tb.tb_frame, 1)[0] + self.assertEqual(line, raise_src.strip()) + + s = io.StringIO() + print_tb(tb, 1, s) + self.assertTrue(s.getvalue().endswith(raise_src)) + else: + raise AssertionError("This ought to be impossible") + + def testTraceback(self): + files = {TESTMOD + ".py": (NOW, raise_src)} + self.doTest(None, files, TESTMOD, call=self.doTraceback) + + @unittest.skipIf(os_helper.TESTFN_UNENCODABLE is None, + "need an unencodable filename") + def testUnencodable(self): + filename = os_helper.TESTFN_UNENCODABLE + ".zip" + self.addCleanup(os_helper.unlink, filename) + with ZipFile(filename, "w") as z: + zinfo = ZipInfo(TESTMOD + ".py", time.localtime(NOW)) + zinfo.compress_type = self.compression + z.writestr(zinfo, test_src) + spec = zipimport.zipimporter(filename).find_spec(TESTMOD) + mod = importlib.util.module_from_spec(spec) + spec.loader.exec_module(mod) + + def testBytesPath(self): + filename = os_helper.TESTFN + ".zip" + self.addCleanup(os_helper.unlink, filename) + with ZipFile(filename, "w") as z: + zinfo = ZipInfo(TESTMOD + ".py", time.localtime(NOW)) + zinfo.compress_type = self.compression + z.writestr(zinfo, test_src) + + zipimport.zipimporter(filename) + zipimport.zipimporter(os.fsencode(filename)) + with self.assertRaises(TypeError): + zipimport.zipimporter(bytearray(os.fsencode(filename))) + with self.assertRaises(TypeError): + zipimport.zipimporter(memoryview(os.fsencode(filename))) + + def testComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, comment=b"comment") + + def testBeginningCruftAndComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, stuff=b"cruft" * 64, comment=b"hi") + + def testLargestPossibleComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, comment=b"c" * ((1 << 16) - 1)) + + +@support.requires_zlib() +class CompressedZipImportTestCase(UncompressedZipImportTestCase): + compression = ZIP_DEFLATED + + +class BadFileZipImportTestCase(unittest.TestCase): + def assertZipFailure(self, filename): + self.assertRaises(zipimport.ZipImportError, + zipimport.zipimporter, filename) + + def testNoFile(self): + self.assertZipFailure('AdfjdkFJKDFJjdklfjs') + + def testEmptyFilename(self): + self.assertZipFailure('') + + def testBadArgs(self): + self.assertRaises(TypeError, zipimport.zipimporter, None) + self.assertRaises(TypeError, zipimport.zipimporter, TESTMOD, kwd=None) + self.assertRaises(TypeError, zipimport.zipimporter, + list(os.fsencode(TESTMOD))) + + def testFilenameTooLong(self): + self.assertZipFailure('A' * 33000) + + def testEmptyFile(self): + os_helper.unlink(TESTMOD) + os_helper.create_empty_file(TESTMOD) + self.assertZipFailure(TESTMOD) + + def testFileUnreadable(self): + os_helper.unlink(TESTMOD) + fd = os.open(TESTMOD, os.O_CREAT, 000) + try: + os.close(fd) + + with self.assertRaises(zipimport.ZipImportError) as cm: + zipimport.zipimporter(TESTMOD) + finally: + # If we leave "the read-only bit" set on Windows, nothing can + # delete TESTMOD, and later tests suffer bogus failures. + os.chmod(TESTMOD, 0o666) + os_helper.unlink(TESTMOD) + + def testNotZipFile(self): + os_helper.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write('a' * 22) + fp.close() + self.assertZipFailure(TESTMOD) + + # XXX: disabled until this works on Big-endian machines + def _testBogusZipFile(self): + os_helper.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write(struct.pack('=I', 0x06054B50)) + fp.write('a' * 18) + fp.close() + z = zipimport.zipimporter(TESTMOD) + + try: + with warnings.catch_warnings(): + warnings.simplefilter("ignore", DeprecationWarning) + self.assertRaises(TypeError, z.load_module, None) + self.assertRaises(TypeError, z.find_module, None) + self.assertRaises(TypeError, z.find_spec, None) + self.assertRaises(TypeError, z.exec_module, None) + self.assertRaises(TypeError, z.is_package, None) + self.assertRaises(TypeError, z.get_code, None) + self.assertRaises(TypeError, z.get_data, None) + self.assertRaises(TypeError, z.get_source, None) + + error = zipimport.ZipImportError + self.assertIsNone(z.find_module('abc')) + self.assertIsNone(z.find_spec('abc')) + + with warnings.catch_warnings(): + warnings.simplefilter("ignore", DeprecationWarning) + self.assertRaises(error, z.load_module, 'abc') + self.assertRaises(error, z.get_code, 'abc') + self.assertRaises(OSError, z.get_data, 'abc') + self.assertRaises(error, z.get_source, 'abc') + self.assertRaises(error, z.is_package, 'abc') + finally: + zipimport._zip_directory_cache.clear() + + +def tearDownModule(): + os_helper.unlink(TESTMOD) + + +if __name__ == "__main__": + unittest.main() diff --git a/cpython-tests/3.10/zip_cp437_header.zip b/cpython-tests/3.10/zip_cp437_header.zip new file mode 100644 index 0000000..f7c6cf1 Binary files /dev/null and b/cpython-tests/3.10/zip_cp437_header.zip differ diff --git a/cpython-tests/3.10/zipdir.zip b/cpython-tests/3.10/zipdir.zip new file mode 100644 index 0000000..ac21d7a Binary files /dev/null and b/cpython-tests/3.10/zipdir.zip differ diff --git a/cpython-tests/3.10/ziptestdata/README.md b/cpython-tests/3.10/ziptestdata/README.md new file mode 100644 index 0000000..6b9147d --- /dev/null +++ b/cpython-tests/3.10/ziptestdata/README.md @@ -0,0 +1,35 @@ +# Test data for `test_zipfile` + +The test executables in this directory are created manually from header.sh and +the `testdata_module_inside_zip.py` file. You must have infozip's zip utility +installed (`apt install zip` on Debian). + +## Purpose + +These are used to test executable files with an appended zipfile, in a scenario +where the executable is _not_ a Python interpreter itself so our automatic +zipimport machinery (that'd look for `__main__.py`) is not being used. + +## Updating the test executables + +If you update header.sh or the testdata_module_inside_zip.py file, rerun the +commands below. These are expected to be rarely changed, if ever. + +### Standard old format (2.0) zip file + +``` +zip -0 zip2.zip testdata_module_inside_zip.py +cat header.sh zip2.zip >exe_with_zip +rm zip2.zip +``` + +### Modern format (4.5) zip64 file + +Redirecting from stdin forces infozip's zip tool to create a zip64. + +``` +zip -0 zip64.zip +cat header.sh zip64.zip >exe_with_z64 +rm zip64.zip +``` + diff --git a/cpython-tests/3.10/ziptestdata/exe_with_z64 b/cpython-tests/3.10/ziptestdata/exe_with_z64 new file mode 100644 index 0000000..82b03cf Binary files /dev/null and b/cpython-tests/3.10/ziptestdata/exe_with_z64 differ diff --git a/cpython-tests/3.10/ziptestdata/exe_with_zip b/cpython-tests/3.10/ziptestdata/exe_with_zip new file mode 100644 index 0000000..c833cdf Binary files /dev/null and b/cpython-tests/3.10/ziptestdata/exe_with_zip differ diff --git a/cpython-tests/3.10/ziptestdata/header.sh b/cpython-tests/3.10/ziptestdata/header.sh new file mode 100644 index 0000000..52dc91a --- /dev/null +++ b/cpython-tests/3.10/ziptestdata/header.sh @@ -0,0 +1,24 @@ +#!/bin/bash +INTERPRETER_UNDER_TEST="$1" +if [[ ! -x "${INTERPRETER_UNDER_TEST}" ]]; then + echo "Interpreter must be the command line argument." + exit 4 +fi +EXECUTABLE="$0" exec "${INTERPRETER_UNDER_TEST}" -E - <e|f"g?h*i', ','), r'a\b,c_d_e_f_g_h_i') + self.assertEqual(san('../../foo../../ba..r', '/'), r'foo/ba..r') + + def test_extract_hackers_arcnames_common_cases(self): + common_hacknames = [ + ('../foo/bar', 'foo/bar'), + ('foo/../bar', 'foo/bar'), + ('foo/../../bar', 'foo/bar'), + ('foo/bar/..', 'foo/bar'), + ('./../foo/bar', 'foo/bar'), + ('/foo/bar', 'foo/bar'), + ('/foo/../bar', 'foo/bar'), + ('/foo/../../bar', 'foo/bar'), + ] + self._test_extract_hackers_arcnames(common_hacknames) + + @unittest.skipIf(os.path.sep != '\\', 'Requires \\ as path separator.') + def test_extract_hackers_arcnames_windows_only(self): + """Test combination of path fixing and windows name sanitization.""" + windows_hacknames = [ + (r'..\foo\bar', 'foo/bar'), + (r'..\/foo\/bar', 'foo/bar'), + (r'foo/\..\/bar', 'foo/bar'), + (r'foo\/../\bar', 'foo/bar'), + (r'C:foo/bar', 'foo/bar'), + (r'C:/foo/bar', 'foo/bar'), + (r'C://foo/bar', 'foo/bar'), + (r'C:\foo\bar', 'foo/bar'), + (r'//conky/mountpoint/foo/bar', 'foo/bar'), + (r'\\conky\mountpoint\foo\bar', 'foo/bar'), + (r'///conky/mountpoint/foo/bar', 'mountpoint/foo/bar'), + (r'\\\conky\mountpoint\foo\bar', 'mountpoint/foo/bar'), + (r'//conky//mountpoint/foo/bar', 'mountpoint/foo/bar'), + (r'\\conky\\mountpoint\foo\bar', 'mountpoint/foo/bar'), + (r'//?/C:/foo/bar', 'foo/bar'), + (r'\\?\C:\foo\bar', 'foo/bar'), + (r'C:/../C:/foo/bar', 'C_/foo/bar'), + (r'a:b\ce|f"g?h*i', 'b/c_d_e_f_g_h_i'), + ('../../foo../../ba..r', 'foo/ba..r'), + ] + self._test_extract_hackers_arcnames(windows_hacknames) + + @unittest.skipIf(os.path.sep != '/', r'Requires / as path separator.') + def test_extract_hackers_arcnames_posix_only(self): + posix_hacknames = [ + ('//foo/bar', 'foo/bar'), + ('../../foo../../ba..r', 'foo../ba..r'), + (r'foo/..\bar', r'foo/..\bar'), + ] + self._test_extract_hackers_arcnames(posix_hacknames) + + def _test_extract_hackers_arcnames(self, hacknames): + for arcname, fixedname in hacknames: + content = b'foobar' + arcname.encode() + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipfp: + zinfo = zipfile.ZipInfo() + # preserve backslashes + zinfo.filename = arcname + zinfo.external_attr = 0o600 << 16 + zipfp.writestr(zinfo, content) + + arcname = arcname.replace(os.sep, "/") + targetpath = os.path.join('target', 'subdir', 'subsub') + correctfile = os.path.join(targetpath, *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname, targetpath) + self.assertEqual(writtenfile, correctfile, + msg='extract %r: %r != %r' % + (arcname, writtenfile, correctfile)) + self.check_file(correctfile, content) + rmtree('target') + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall(targetpath) + self.check_file(correctfile, content) + rmtree('target') + + correctfile = os.path.join(os.getcwd(), *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname) + self.assertEqual(writtenfile, correctfile, + msg="extract %r" % arcname) + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall() + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + unlink(TESTFN2) + + +class OtherTests(unittest.TestCase): + def test_open_via_zip_info(self): + # Create the ZIP archive + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED, low_level=True) as zipfp: + zipfp.writestr("name", "foo") + with self.assertWarns(UserWarning): + zipfp.writestr("name", "bar") + self.assertEqual(zipfp.namelist(), ["name"] * 2) + + with zipfile2.ZipFile(TESTFN2, "r", low_level=True) as zipfp: + infos = zipfp.infolist() + data = b"" + for info in infos: + with zipfp.open(info) as zipopen: + data += zipopen.read() + self.assertIn(data, {b"foobar", b"barfoo"}) + data = b"" + for info in infos: + data += zipfp.read(info) + self.assertIn(data, {b"foobar", b"barfoo"}) + + def test_writestr_extended_local_header_issue1202(self): + with zipfile2.ZipFile(TESTFN2, 'w') as orig_zip: + for data in 'abcdefghijklmnop': + zinfo = zipfile.ZipInfo(data) + zinfo.flag_bits |= zipfile._MASK_USE_DATA_DESCRIPTOR # Include an extended local header. + orig_zip.writestr(zinfo, data) + + def test_close(self): + """Check that the zipfile is closed after the 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + def test_close_on_exception(self): + """Check that the zipfile is closed if an exception is raised in the + 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + + try: + with zipfile2.ZipFile(TESTFN2, "r") as zipfp2: + raise zipfile.BadZipFile() + except zipfile.BadZipFile: + self.assertIsNone(zipfp2.fp, 'zipfp is not closed') + + def test_unsupported_version(self): + # File has an extract_version of 120 + data = (b'PK\x03\x04x\x00\x00\x00\x00\x00!p\xa1@\x00\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x01\x00\x00\x00xPK\x01\x02x\x03x\x00\x00\x00\x00' + b'\x00!p\xa1@\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x01\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00\x00xPK\x05\x06' + b'\x00\x00\x00\x00\x01\x00\x01\x00/\x00\x00\x00\x1f\x00\x00\x00\x00\x00') + + self.assertRaises(NotImplementedError, zipfile2.ZipFile, + io.BytesIO(data), 'r') + + @requires_zlib() + def test_read_unicode_filenames(self): + # bug #10801 + fname = findfile('zip_cp437_header.zip') + with zipfile2.ZipFile(fname) as zipfp: + for name in zipfp.namelist(): + zipfp.open(name).close() + + def test_write_unicode_filenames(self): + with zipfile2.ZipFile(TESTFN, "w") as zf: + zf.writestr("foo.txt", "Test for unicode filename") + zf.writestr("\xf6.txt", "Test for unicode filename") + self.assertIsInstance(zf.infolist()[0].filename, str) + + with zipfile2.ZipFile(TESTFN, "r") as zf: + self.assertEqual(zf.filelist[0].filename, "foo.txt") + self.assertEqual(zf.filelist[1].filename, "\xf6.txt") + + def test_read_after_write_unicode_filenames(self): + with zipfile2.ZipFile(TESTFN2, 'w') as zipfp: + zipfp.writestr('приклад', b'sample') + self.assertEqual(zipfp.read('приклад'), b'sample') + + def test_exclusive_create_zip_file(self): + """Test exclusive creating a new zipfile.""" + unlink(TESTFN2) + filename = 'testfile.txt' + content = b'hello, world. this is some content.' + with zipfile2.ZipFile(TESTFN2, "x", zipfile.ZIP_STORED) as zipfp: + zipfp.writestr(filename, content) + with self.assertRaises(FileExistsError): + zipfile2.ZipFile(TESTFN2, "x", zipfile.ZIP_STORED) + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertEqual(zipfp.namelist(), [filename]) + self.assertEqual(zipfp.read(filename), content) + + def test_create_non_existent_file_for_append(self): + if os.path.exists(TESTFN): + os.unlink(TESTFN) + + filename = 'testfile.txt' + content = b'hello, world. this is some content.' + + try: + with zipfile2.ZipFile(TESTFN, 'a') as zf: + zf.writestr(filename, content) + except OSError: + self.fail('Could not append data to a non-existent zip file.') + + self.assertTrue(os.path.exists(TESTFN)) + + with zipfile2.ZipFile(TESTFN, 'r') as zf: + self.assertEqual(zf.read(filename), content) + + def test_close_erroneous_file(self): + # This test checks that the ZipFile constructor closes the file object + # it opens if there's an error in the file. If it doesn't, the + # traceback holds a reference to the ZipFile object and, indirectly, + # the file object. + # On Windows, this causes the os.unlink() call to fail because the + # underlying file is still open. This is SF bug #412214. + # + with open(TESTFN, "w", encoding="utf-8") as fp: + fp.write("this is not a legal zip file\n") + try: + zf = zipfile2.ZipFile(TESTFN) + except zipfile.BadZipFile: + pass + + def test_is_zip_erroneous_file(self): + """Check that is_zipfile() correctly identifies non-zip files.""" + # - passing a filename + with open(TESTFN, "w", encoding='utf-8') as fp: + fp.write("this is not a legal zip file\n") + self.assertFalse(zipfile.is_zipfile(TESTFN)) + # - passing a path-like object + self.assertFalse(zipfile.is_zipfile(pathlib.Path(TESTFN))) + # - passing a file object + with open(TESTFN, "rb") as fp: + self.assertFalse(zipfile.is_zipfile(fp)) + # - passing a file-like object + fp = io.BytesIO() + fp.write(b"this is not a legal zip file\n") + self.assertFalse(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + self.assertFalse(zipfile.is_zipfile(fp)) + + def test_damaged_zipfile(self): + """Check that zipfiles with missing bytes at the end raise BadZipFile.""" + # - Create a valid zip file + fp = io.BytesIO() + with zipfile2.ZipFile(fp, mode="w") as zipf: + zipf.writestr("foo.txt", b"O, for a Muse of Fire!") + zipfiledata = fp.getvalue() + + # - Now create copies of it missing the last N bytes and make sure + # a BadZipFile exception is raised when we try to open it + for N in range(len(zipfiledata)): + fp = io.BytesIO(zipfiledata[:N]) + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, fp) + + def test_is_zip_valid_file(self): + """Check that is_zipfile() correctly identifies zip files.""" + # - passing a filename + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt", b"O, for a Muse of Fire!") + + self.assertTrue(zipfile.is_zipfile(TESTFN)) + # - passing a file object + with open(TESTFN, "rb") as fp: + self.assertTrue(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + zip_contents = fp.read() + # - passing a file-like object + fp = io.BytesIO() + fp.write(zip_contents) + self.assertTrue(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + self.assertTrue(zipfile.is_zipfile(fp)) + + def test_non_existent_file_raises_OSError(self): + # make sure we don't raise an AttributeError when a partially-constructed + # ZipFile instance is finalized; this tests for regression on SF tracker + # bug #403871. + + # The bug we're testing for caused an AttributeError to be raised + # when a ZipFile instance was created for a file that did not + # exist; the .fp member was not initialized but was needed by the + # __del__() method. Since the AttributeError is in the __del__(), + # it is ignored, but the user should be sufficiently annoyed by + # the message on the output that regression will be noticed + # quickly. + self.assertRaises(OSError, zipfile2.ZipFile, TESTFN) + + def test_empty_file_raises_BadZipFile(self): + f = open(TESTFN, 'w', encoding='utf-8') + f.close() + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN) + + with open(TESTFN, 'w', encoding='utf-8') as fp: + fp.write("short file") + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN) + + def test_negative_central_directory_offset_raises_BadZipFile(self): + # Zip file containing an empty EOCD record + buffer = bytearray(b'PK\x05\x06' + b'\0'*18) + + # Set the size of the central directory bytes to become 1, + # causing the central directory offset to become negative + for dirsize in 1, 2**32-1: + buffer[12:16] = struct.pack(' os.path.getsize(TESTFN)) + with zipfile2.ZipFile(TESTFN,mode="r") as zipf: + self.assertEqual(zipf.comment, b"shorter comment") + + def test_unicode_comment(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with self.assertRaises(TypeError): + zipf.comment = "this is an error" + + def test_change_comment_in_empty_archive(self): + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertFalse(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_change_comment_in_nonempty_archive(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertTrue(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_empty_zipfile(self): + # Check that creating a file in 'w' or 'a' mode and closing without + # adding any files to the archives creates a valid empty ZIP file + zipf = zipfile2.ZipFile(TESTFN, mode="w") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except zipfile.BadZipFile: + self.fail("Unable to create empty ZIP file in 'w' mode") + + zipf = zipfile2.ZipFile(TESTFN, mode="a") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except: + self.fail("Unable to create empty ZIP file in 'a' mode") + + def test_open_empty_file(self): + # Issue 1710703: Check that opening a file with less than 22 bytes + # raises a BadZipFile exception (rather than the previously unhelpful + # OSError) + f = open(TESTFN, 'w', encoding='utf-8') + f.close() + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN, 'r') + + def test_create_zipinfo_before_1980(self): + self.assertRaises(ValueError, + zipfile.ZipInfo, 'seventies', (1979, 1, 1, 0, 0, 0)) + + def test_create_empty_zipinfo_repr(self): + """Before bpo-26185, repr() on empty ZipInfo object was failing.""" + zi = zipfile.ZipInfo(filename="empty") + self.assertEqual(repr(zi), "") + + def test_create_empty_zipinfo_default_attributes(self): + """Ensure all required attributes are set.""" + zi = zipfile.ZipInfo() + self.assertEqual(zi.orig_filename, "NoName") + self.assertEqual(zi.filename, "NoName") + self.assertEqual(zi.date_time, (1980, 1, 1, 0, 0, 0)) + self.assertEqual(zi.compress_type, zipfile.ZIP_STORED) + self.assertEqual(zi.comment, b"") + self.assertEqual(zi.extra, b"") + self.assertIn(zi.create_system, (0, 3)) + self.assertEqual(zi.create_version, zipfile.DEFAULT_VERSION) + self.assertEqual(zi.extract_version, zipfile.DEFAULT_VERSION) + self.assertEqual(zi.reserved, 0) + self.assertEqual(zi.flag_bits, 0) + self.assertEqual(zi.volume, 0) + self.assertEqual(zi.internal_attr, 0) + self.assertEqual(zi.external_attr, 0) + + # Before bpo-26185, both were missing + self.assertEqual(zi.file_size, 0) + self.assertEqual(zi.compress_size, 0) + + def test_zipfile_with_short_extra_field(self): + """If an extra field in the header is less than 4 bytes, skip it.""" + zipdata = ( + b'PK\x03\x04\x14\x00\x00\x00\x00\x00\x93\x9b\xad@\x8b\x9e' + b'\xd9\xd3\x01\x00\x00\x00\x01\x00\x00\x00\x03\x00\x03\x00ab' + b'c\x00\x00\x00APK\x01\x02\x14\x03\x14\x00\x00\x00\x00' + b'\x00\x93\x9b\xad@\x8b\x9e\xd9\xd3\x01\x00\x00\x00\x01\x00\x00' + b'\x00\x03\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\xa4\x81\x00' + b'\x00\x00\x00abc\x00\x00PK\x05\x06\x00\x00\x00\x00' + b'\x01\x00\x01\x003\x00\x00\x00%\x00\x00\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(zipdata), 'r') as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertIsNone(zipf.testzip()) + + def test_open_conflicting_handles(self): + # It's only possible to open one writable file handle at a time + msg1 = b"It's fun to charter an accountant!" + msg2 = b"And sail the wide accountant sea" + msg3 = b"To find, explore the funds offshore" + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipf: + with zipf.open('foo', mode='w') as w2: + w2.write(msg1) + with zipf.open('bar', mode='w') as w1: + with self.assertRaises(ValueError): + zipf.open('handle', mode='w') + with self.assertRaises(ValueError): + zipf.open('foo', mode='r') + with self.assertRaises(ValueError): + zipf.writestr('str', 'abcde') + with self.assertRaises(ValueError): + zipf.write(__file__, 'file') + with self.assertRaises(ValueError): + zipf.close() + w1.write(msg2) + with zipf.open('baz', mode='w') as w2: + w2.write(msg3) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipf: + self.assertEqual(zipf.read('foo'), msg1) + self.assertEqual(zipf.read('bar'), msg2) + self.assertEqual(zipf.read('baz'), msg3) + self.assertEqual(zipf.namelist(), ['foo', 'bar', 'baz']) + + def test_seek_tell(self): + # Test seek functionality + txt = b"Where's Bruce?" + bloc = txt.find(b"Bruce") + # Check seek on a file + with zipfile2.ZipFile(TESTFN, "w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(TESTFN, "r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + # Check seek on memory file + data = io.BytesIO() + with zipfile2.ZipFile(data, mode="w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(data, mode="r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + + @requires_bz2() + def test_decompress_without_3rd_party_library(self): + data = b'PK\x05\x06\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + zip_file = io.BytesIO(data) + with zipfile2.ZipFile(zip_file, 'w', compression=zipfile.ZIP_BZIP2) as zf: + zf.writestr('a.txt', b'a') + with mock.patch('zipfile.bz2', None): + with zipfile2.ZipFile(zip_file) as zf: + self.assertRaises(RuntimeError, zf.extract, 'a.txt') + + def tearDown(self): + unlink(TESTFN) + unlink(TESTFN2) + + +class AbstractBadCrcTests: + def test_testzip_with_bad_crc(self): + """Tests that files with bad CRCs return their name from testzip.""" + zipdata = self.zip_with_bad_crc + + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertEqual('afile', zipf.testzip()) + + def test_read_with_bad_crc(self): + """Tests that files with bad CRCs raise a BadZipFile exception when read.""" + zipdata = self.zip_with_bad_crc + + # Using ZipFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + self.assertRaises(zipfile.BadZipFile, zipf.read, 'afile') + + # Using ZipExtFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + self.assertRaises(zipfile.BadZipFile, corrupt_file.read) + + # Same with small reads (in order to exercise the buffering logic) + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + corrupt_file.MIN_READ_SIZE = 2 + with self.assertRaises(zipfile.BadZipFile): + while corrupt_file.read(2): + pass + + +class StoredBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_STORED + zip_with_bad_crc = ( + b'PK\003\004\024\0\0\0\0\0 \213\212;:r' + b'\253\377\f\0\0\0\f\0\0\0\005\0\0\000af' + b'ilehello,AworldP' + b'K\001\002\024\003\024\0\0\0\0\0 \213\212;:' + b'r\253\377\f\0\0\0\f\0\0\0\005\0\0\0\0' + b'\0\0\0\0\0\0\0\200\001\0\0\0\000afi' + b'lePK\005\006\0\0\0\0\001\0\001\0003\000' + b'\0\0/\0\0\0\0\0') + +@requires_zlib() +class DeflateBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_DEFLATED + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00n}\x0c=FA' + b'KE\x10\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\xcbH\xcd\xc9\xc9W(\xcf/\xcaI\xc9\xa0' + b'=\x13\x00PK\x01\x02\x14\x03\x14\x00\x00\x00\x08\x00n' + b'}\x0c=FAKE\x10\x00\x00\x00n\x00\x00\x00\x05' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00' + b'\x00afilePK\x05\x06\x00\x00\x00\x00\x01\x00' + b'\x01\x003\x00\x00\x003\x00\x00\x00\x00\x00') + +@requires_bz2() +class Bzip2BadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_BZIP2 + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0c\x00nu\x0c=FA' + b'KE8\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ileBZh91AY&SY\xd4\xa8\xca' + b'\x7f\x00\x00\x0f\x11\x80@\x00\x06D\x90\x80 \x00 \xa5' + b'P\xd9!\x03\x03\x13\x13\x13\x89\xa9\xa9\xc2u5:\x9f' + b'\x8b\xb9"\x9c(HjTe?\x80PK\x01\x02\x14' + b'\x03\x14\x03\x00\x00\x0c\x00nu\x0c=FAKE8' + b'\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x00 \x80\x80\x81\x00\x00\x00\x00afilePK' + b'\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00\x00[\x00' + b'\x00\x00\x00\x00') + +@requires_lzma() +class LzmaBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_LZMA + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\t\x04\x05\x00]\x00\x00\x00\x04\x004\x19I' + b'\xee\x8d\xe9\x17\x89:3`\tq!.8\x00PK' + b'\x01\x02\x14\x03\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00 \x80\x80\x81\x00\x00\x00\x00afil' + b'ePK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00' + b'\x00>\x00\x00\x00\x00\x00') + + +class DecryptionTests(unittest.TestCase): + """Check that ZIP decryption works. Since the library does not + support encryption at the moment, we use a pre-generated encrypted + ZIP file.""" + + data = ( + b'PK\x03\x04\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00\x1a\x00' + b'\x00\x00\x08\x00\x00\x00test.txt\xfa\x10\xa0gly|\xfa-\xc5\xc0=\xf9y' + b'\x18\xe0\xa8r\xb3Z}Lg\xbc\xae\xf9|\x9b\x19\xe4\x8b\xba\xbb)\x8c\xb0\xdbl' + b'PK\x01\x02\x14\x00\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00' + b'\x1a\x00\x00\x00\x08\x00\x00\x00\x00\x00\x00\x00\x01\x00 \x00\xb6\x81' + b'\x00\x00\x00\x00test.txtPK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x006\x00' + b'\x00\x00L\x00\x00\x00\x00\x00' ) + data2 = ( + b'PK\x03\x04\x14\x00\t\x00\x08\x00\xcf}38xu\xaa\xb2\x14\x00\x00\x00\x00\x02' + b'\x00\x00\x04\x00\x15\x00zeroUT\t\x00\x03\xd6\x8b\x92G\xda\x8b\x92GUx\x04' + b'\x00\xe8\x03\xe8\x03\xc7", "exec") +raise_src = 'def do_raise(): raise TypeError\n' + +def make_pyc(co, mtime, size): + data = marshal.dumps(co) + pyc = (importlib.util.MAGIC_NUMBER + + struct.pack("", "exec"), NOW, len(src)) + files = {TESTMOD + pyc_ext: (NOW, pyc), + "some.data": (NOW, "some data")} + self.doTest(pyc_ext, files, TESTMOD) + + def testDefaultOptimizationLevel(self): + # zipimport should use the default optimization level (#28131) + src = """if 1: # indent hack + def test(val): + assert(val) + return val\n""" + files = {TESTMOD + '.py': (NOW, src)} + self.makeZip(files) + sys.path.insert(0, TEMP_ZIP) + mod = importlib.import_module(TESTMOD) + self.assertEqual(mod.test(1), 1) + if __debug__: + self.assertRaises(AssertionError, mod.test, False) + else: + self.assertEqual(mod.test(0), 0) + + def testImport_WithStuff(self): + # try importing from a zipfile which contains additional + # stuff at the beginning of the file + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, + stuff=b"Some Stuff"*31) + + def assertModuleSource(self, module): + self.assertEqual(inspect.getsource(module), test_src) + + def testGetSource(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, call=self.assertModuleSource) + + def testGetCompiledSource(self): + pyc = make_pyc(compile(test_src, "", "exec"), NOW, len(test_src)) + files = {TESTMOD + ".py": (NOW, test_src), + TESTMOD + pyc_ext: (NOW, pyc)} + self.doTest(pyc_ext, files, TESTMOD, call=self.assertModuleSource) + + def runDoctest(self, callback): + files = {TESTMOD + ".py": (NOW, test_src), + "xyz.txt": (NOW, ">>> log.append(True)\n")} + self.doTest(".py", files, TESTMOD, call=callback) + + def doDoctestFile(self, module): + log = [] + old_master, doctest.master = doctest.master, None + try: + doctest.testfile( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ) + finally: + doctest.master = old_master + self.assertEqual(log,[True]) + + def testDoctestFile(self): + self.runDoctest(self.doDoctestFile) + + def doDoctestSuite(self, module): + log = [] + doctest.DocFileTest( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ).run() + self.assertEqual(log,[True]) + + def testDoctestSuite(self): + self.runDoctest(self.doDoctestSuite) + + def doTraceback(self, module): + try: + module.do_raise() + except Exception as e: + tb = e.__traceback__.tb_next + + f,lno,n,line = extract_tb(tb, 1)[0] + self.assertEqual(line, raise_src.strip()) + + f,lno,n,line = extract_stack(tb.tb_frame, 1)[0] + self.assertEqual(line, raise_src.strip()) + + s = io.StringIO() + print_tb(tb, 1, s) + self.assertTrue(s.getvalue().endswith( + ' def do_raise(): raise TypeError\n' + '' if support.has_no_debug_ranges() else + ' ^^^^^^^^^^^^^^^\n' + )) + else: + raise AssertionError("This ought to be impossible") + + def testTraceback(self): + files = {TESTMOD + ".py": (NOW, raise_src)} + self.doTest(None, files, TESTMOD, call=self.doTraceback) + + @unittest.skipIf(os_helper.TESTFN_UNENCODABLE is None, + "need an unencodable filename") + def testUnencodable(self): + filename = os_helper.TESTFN_UNENCODABLE + ".zip" + self.addCleanup(os_helper.unlink, filename) + with ZipFile(filename, "w") as z: + zinfo = ZipInfo(TESTMOD + ".py", time.localtime(NOW)) + zinfo.compress_type = self.compression + z.writestr(zinfo, test_src) + spec = zipimport.zipimporter(filename).find_spec(TESTMOD) + mod = importlib.util.module_from_spec(spec) + spec.loader.exec_module(mod) + + def testBytesPath(self): + filename = os_helper.TESTFN + ".zip" + self.addCleanup(os_helper.unlink, filename) + with ZipFile(filename, "w") as z: + zinfo = ZipInfo(TESTMOD + ".py", time.localtime(NOW)) + zinfo.compress_type = self.compression + z.writestr(zinfo, test_src) + + zipimport.zipimporter(filename) + with self.assertRaises(TypeError): + zipimport.zipimporter(os.fsencode(filename)) + with self.assertRaises(TypeError): + zipimport.zipimporter(bytearray(os.fsencode(filename))) + with self.assertRaises(TypeError): + zipimport.zipimporter(memoryview(os.fsencode(filename))) + + def testComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, comment=b"comment") + + def testBeginningCruftAndComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, stuff=b"cruft" * 64, comment=b"hi") + + def testLargestPossibleComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, comment=b"c" * ((1 << 16) - 1)) + + +@support.requires_zlib() +class CompressedZipImportTestCase(UncompressedZipImportTestCase): + compression = ZIP_DEFLATED + + +class BadFileZipImportTestCase(unittest.TestCase): + def assertZipFailure(self, filename): + self.assertRaises(zipimport.ZipImportError, + zipimport.zipimporter, filename) + + def testNoFile(self): + self.assertZipFailure('AdfjdkFJKDFJjdklfjs') + + def testEmptyFilename(self): + self.assertZipFailure('') + + def testBadArgs(self): + self.assertRaises(TypeError, zipimport.zipimporter, None) + self.assertRaises(TypeError, zipimport.zipimporter, TESTMOD, kwd=None) + self.assertRaises(TypeError, zipimport.zipimporter, + list(os.fsencode(TESTMOD))) + + def testFilenameTooLong(self): + self.assertZipFailure('A' * 33000) + + def testEmptyFile(self): + os_helper.unlink(TESTMOD) + os_helper.create_empty_file(TESTMOD) + self.assertZipFailure(TESTMOD) + + @unittest.skipIf(support.is_wasi, "mode 000 not supported.") + def testFileUnreadable(self): + os_helper.unlink(TESTMOD) + fd = os.open(TESTMOD, os.O_CREAT, 000) + try: + os.close(fd) + + with self.assertRaises(zipimport.ZipImportError) as cm: + zipimport.zipimporter(TESTMOD) + finally: + # If we leave "the read-only bit" set on Windows, nothing can + # delete TESTMOD, and later tests suffer bogus failures. + os.chmod(TESTMOD, 0o666) + os_helper.unlink(TESTMOD) + + def testNotZipFile(self): + os_helper.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write('a' * 22) + fp.close() + self.assertZipFailure(TESTMOD) + + # XXX: disabled until this works on Big-endian machines + def _testBogusZipFile(self): + os_helper.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write(struct.pack('=I', 0x06054B50)) + fp.write('a' * 18) + fp.close() + z = zipimport.zipimporter(TESTMOD) + + try: + with warnings.catch_warnings(): + warnings.simplefilter("ignore", DeprecationWarning) + self.assertRaises(TypeError, z.load_module, None) + self.assertRaises(TypeError, z.find_module, None) + self.assertRaises(TypeError, z.find_spec, None) + self.assertRaises(TypeError, z.exec_module, None) + self.assertRaises(TypeError, z.is_package, None) + self.assertRaises(TypeError, z.get_code, None) + self.assertRaises(TypeError, z.get_data, None) + self.assertRaises(TypeError, z.get_source, None) + + error = zipimport.ZipImportError + self.assertIsNone(z.find_module('abc')) + self.assertIsNone(z.find_spec('abc')) + + with warnings.catch_warnings(): + warnings.simplefilter("ignore", DeprecationWarning) + self.assertRaises(error, z.load_module, 'abc') + self.assertRaises(error, z.get_code, 'abc') + self.assertRaises(OSError, z.get_data, 'abc') + self.assertRaises(error, z.get_source, 'abc') + self.assertRaises(error, z.is_package, 'abc') + finally: + zipimport._zip_directory_cache.clear() + + +def tearDownModule(): + os_helper.unlink(TESTMOD) + + +if __name__ == "__main__": + unittest.main() diff --git a/cpython-tests/3.11/zip_cp437_header.zip b/cpython-tests/3.11/zip_cp437_header.zip new file mode 100644 index 0000000..f7c6cf1 Binary files /dev/null and b/cpython-tests/3.11/zip_cp437_header.zip differ diff --git a/cpython-tests/3.11/zipdir.zip b/cpython-tests/3.11/zipdir.zip new file mode 100644 index 0000000..ac21d7a Binary files /dev/null and b/cpython-tests/3.11/zipdir.zip differ diff --git a/cpython-tests/3.11/ziptestdata/README.md b/cpython-tests/3.11/ziptestdata/README.md new file mode 100644 index 0000000..6b9147d --- /dev/null +++ b/cpython-tests/3.11/ziptestdata/README.md @@ -0,0 +1,35 @@ +# Test data for `test_zipfile` + +The test executables in this directory are created manually from header.sh and +the `testdata_module_inside_zip.py` file. You must have infozip's zip utility +installed (`apt install zip` on Debian). + +## Purpose + +These are used to test executable files with an appended zipfile, in a scenario +where the executable is _not_ a Python interpreter itself so our automatic +zipimport machinery (that'd look for `__main__.py`) is not being used. + +## Updating the test executables + +If you update header.sh or the testdata_module_inside_zip.py file, rerun the +commands below. These are expected to be rarely changed, if ever. + +### Standard old format (2.0) zip file + +``` +zip -0 zip2.zip testdata_module_inside_zip.py +cat header.sh zip2.zip >exe_with_zip +rm zip2.zip +``` + +### Modern format (4.5) zip64 file + +Redirecting from stdin forces infozip's zip tool to create a zip64. + +``` +zip -0 zip64.zip +cat header.sh zip64.zip >exe_with_z64 +rm zip64.zip +``` + diff --git a/cpython-tests/3.11/ziptestdata/exe_with_z64 b/cpython-tests/3.11/ziptestdata/exe_with_z64 new file mode 100644 index 0000000..82b03cf Binary files /dev/null and b/cpython-tests/3.11/ziptestdata/exe_with_z64 differ diff --git a/cpython-tests/3.11/ziptestdata/exe_with_zip b/cpython-tests/3.11/ziptestdata/exe_with_zip new file mode 100644 index 0000000..c833cdf Binary files /dev/null and b/cpython-tests/3.11/ziptestdata/exe_with_zip differ diff --git a/cpython-tests/3.11/ziptestdata/header.sh b/cpython-tests/3.11/ziptestdata/header.sh new file mode 100644 index 0000000..52dc91a --- /dev/null +++ b/cpython-tests/3.11/ziptestdata/header.sh @@ -0,0 +1,24 @@ +#!/bin/bash +INTERPRETER_UNDER_TEST="$1" +if [[ ! -x "${INTERPRETER_UNDER_TEST}" ]]; then + echo "Interpreter must be the command line argument." + exit 4 +fi +EXECUTABLE="$0" exec "${INTERPRETER_UNDER_TEST}" -E - <>> items = Counter(range(20)) + >>> items.count + 0 + >>> values = list(items) + >>> items.count + 20 + """ + + def __init__(self, i): + self.count = 0 + self.iter = zip(itertools.count(1), i) + + def __iter__(self): + return self + + def __next__(self): + self.count, result = next(self.iter) + return result + + +# from more_itertools v8.13.0 +def always_iterable(obj, base_type=(str, bytes)): + if obj is None: + return iter(()) + + if (base_type is not None) and isinstance(obj, base_type): + return iter((obj,)) + + try: + return iter(obj) + except TypeError: + return iter((obj,)) + + +# from more_itertools v9.0.0 +def consume(iterator, n=None): + """Advance *iterable* by *n* steps. If *n* is ``None``, consume it + entirely. + Efficiently exhausts an iterator without returning values. Defaults to + consuming the whole iterator, but an optional second argument may be + provided to limit consumption. + >>> i = (x for x in range(10)) + >>> next(i) + 0 + >>> consume(i, 3) + >>> next(i) + 4 + >>> consume(i) + >>> next(i) + Traceback (most recent call last): + File "", line 1, in + StopIteration + If the iterator has fewer items remaining than the provided limit, the + whole iterator will be consumed. + >>> i = (x for x in range(3)) + >>> consume(i, 5) + >>> next(i) + Traceback (most recent call last): + File "", line 1, in + StopIteration + """ + # Use functions that consume iterators at C speed. + if n is None: + # feed the entire iterator into a zero-length deque + deque(iterator, maxlen=0) + else: + # advance to the empty slice starting at position n + next(islice(iterator, n, n), None) diff --git a/cpython-tests/3.12/test_zipfile/_path/_support.py b/cpython-tests/3.12/test_zipfile/_path/_support.py new file mode 100644 index 0000000..1afdf3b --- /dev/null +++ b/cpython-tests/3.12/test_zipfile/_path/_support.py @@ -0,0 +1,9 @@ +import importlib +import unittest + + +def import_or_skip(name): + try: + return importlib.import_module(name) + except ImportError: # pragma: no cover + raise unittest.SkipTest(f'Unable to import {name}') diff --git a/cpython-tests/3.12/test_zipfile/_path/_test_params.py b/cpython-tests/3.12/test_zipfile/_path/_test_params.py new file mode 100644 index 0000000..bc95b4e --- /dev/null +++ b/cpython-tests/3.12/test_zipfile/_path/_test_params.py @@ -0,0 +1,39 @@ +import types +import functools + +from ._itertools import always_iterable + + +def parameterize(names, value_groups): + """ + Decorate a test method to run it as a set of subtests. + + Modeled after pytest.parametrize. + """ + + def decorator(func): + @functools.wraps(func) + def wrapped(self): + for values in value_groups: + resolved = map(Invoked.eval, always_iterable(values)) + params = dict(zip(always_iterable(names), resolved)) + with self.subTest(**params): + func(self, **params) + + return wrapped + + return decorator + + +class Invoked(types.SimpleNamespace): + """ + Wrap a function to be invoked for each usage. + """ + + @classmethod + def wrap(cls, func): + return cls(func=func) + + @classmethod + def eval(cls, cand): + return cand.func() if isinstance(cand, cls) else cand diff --git a/cpython-tests/3.12/test_zipfile/_path/test_path.py b/cpython-tests/3.12/test_zipfile/_path/test_path.py new file mode 100644 index 0000000..616c4e8 --- /dev/null +++ b/cpython-tests/3.12/test_zipfile/_path/test_path.py @@ -0,0 +1,676 @@ +import io +import itertools +import contextlib +import pathlib +import pickle +import sys +import time +import unittest +import zipfile + +from test.support.os_helper import temp_dir, FakePath + +from ._functools import compose +from ._itertools import Counter + +from ._test_params import parameterize, Invoked + + +class jaraco: + class itertools: + Counter = Counter + + +def add_dirs(zf): + """ + Given a writable zip file zf, inject directory entries for + any directories implied by the presence of children. + """ + for name in zipfile.CompleteDirs._implied_dirs(zf.namelist()): + zf.writestr(name, b"") + return zf + + +def build_alpharep_fixture(): + """ + Create a zip file with this structure: + + . + ├── a.txt + ├── b + │ ├── c.txt + │ ├── d + │ │ └── e.txt + │ └── f.txt + ├── g + │ └── h + │ └── i.txt + └── j + ├── k.bin + ├── l.baz + └── m.bar + + This fixture has the following key characteristics: + + - a file at the root (a) + - a file two levels deep (b/d/e) + - multiple files in a directory (b/c, b/f) + - a directory containing only a directory (g/h) + - a directory with files of different extensions (j/klm) + + "alpha" because it uses alphabet + "rep" because it's a representative example + """ + data = io.BytesIO() + zf = zipfile.ZipFile(data, "w") + zf.writestr("a.txt", b"content of a") + zf.writestr("b/c.txt", b"content of c") + zf.writestr("b/d/e.txt", b"content of e") + zf.writestr("b/f.txt", b"content of f") + zf.writestr("g/h/i.txt", b"content of i") + zf.writestr("j/k.bin", b"content of k") + zf.writestr("j/l.baz", b"content of l") + zf.writestr("j/m.bar", b"content of m") + zf.filename = "alpharep.zip" + return zf + + +alpharep_generators = [ + Invoked.wrap(build_alpharep_fixture), + Invoked.wrap(compose(add_dirs, build_alpharep_fixture)), +] + +pass_alpharep = parameterize(['alpharep'], alpharep_generators) + + +class TestPath(unittest.TestCase): + def setUp(self): + self.fixtures = contextlib.ExitStack() + self.addCleanup(self.fixtures.close) + + def zipfile_ondisk(self, alpharep): + tmpdir = pathlib.Path(self.fixtures.enter_context(temp_dir())) + buffer = alpharep.fp + alpharep.close() + path = tmpdir / alpharep.filename + with path.open("wb") as strm: + strm.write(buffer.getvalue()) + return path + + @pass_alpharep + def test_iterdir_and_types(self, alpharep): + root = zipfile.Path(alpharep) + assert root.is_dir() + a, b, g, j = root.iterdir() + assert a.is_file() + assert b.is_dir() + assert g.is_dir() + c, f, d = b.iterdir() + assert c.is_file() and f.is_file() + (e,) = d.iterdir() + assert e.is_file() + (h,) = g.iterdir() + (i,) = h.iterdir() + assert i.is_file() + + @pass_alpharep + def test_is_file_missing(self, alpharep): + root = zipfile.Path(alpharep) + assert not root.joinpath('missing.txt').is_file() + + @pass_alpharep + def test_iterdir_on_file(self, alpharep): + root = zipfile.Path(alpharep) + a, b, g, j = root.iterdir() + with self.assertRaises(ValueError): + a.iterdir() + + @pass_alpharep + def test_subdir_is_dir(self, alpharep): + root = zipfile.Path(alpharep) + assert (root / 'b').is_dir() + assert (root / 'b/').is_dir() + assert (root / 'g').is_dir() + assert (root / 'g/').is_dir() + + @pass_alpharep + def test_open(self, alpharep): + root = zipfile.Path(alpharep) + a, b, g, j = root.iterdir() + with a.open(encoding="utf-8") as strm: + data = strm.read() + self.assertEqual(data, "content of a") + with a.open('r', "utf-8") as strm: # not a kw, no gh-101144 TypeError + data = strm.read() + self.assertEqual(data, "content of a") + + def test_open_encoding_utf16(self): + in_memory_file = io.BytesIO() + zf = zipfile.ZipFile(in_memory_file, "w") + zf.writestr("path/16.txt", "This was utf-16".encode("utf-16")) + zf.filename = "test_open_utf16.zip" + root = zipfile.Path(zf) + (path,) = root.iterdir() + u16 = path.joinpath("16.txt") + with u16.open('r', "utf-16") as strm: + data = strm.read() + assert data == "This was utf-16" + with u16.open(encoding="utf-16") as strm: + data = strm.read() + assert data == "This was utf-16" + + def test_open_encoding_errors(self): + in_memory_file = io.BytesIO() + zf = zipfile.ZipFile(in_memory_file, "w") + zf.writestr("path/bad-utf8.bin", b"invalid utf-8: \xff\xff.") + zf.filename = "test_read_text_encoding_errors.zip" + root = zipfile.Path(zf) + (path,) = root.iterdir() + u16 = path.joinpath("bad-utf8.bin") + + # encoding= as a positional argument for gh-101144. + data = u16.read_text("utf-8", errors="ignore") + assert data == "invalid utf-8: ." + with u16.open("r", "utf-8", errors="surrogateescape") as f: + assert f.read() == "invalid utf-8: \udcff\udcff." + + # encoding= both positional and keyword is an error; gh-101144. + with self.assertRaisesRegex(TypeError, "encoding"): + data = u16.read_text("utf-8", encoding="utf-8") + + # both keyword arguments work. + with u16.open("r", encoding="utf-8", errors="strict") as f: + # error during decoding with wrong codec. + with self.assertRaises(UnicodeDecodeError): + f.read() + + @unittest.skipIf( + not getattr(sys.flags, 'warn_default_encoding', 0), + "Requires warn_default_encoding", + ) + @pass_alpharep + def test_encoding_warnings(self, alpharep): + """EncodingWarning must blame the read_text and open calls.""" + assert sys.flags.warn_default_encoding + root = zipfile.Path(alpharep) + with self.assertWarns(EncodingWarning) as wc: + root.joinpath("a.txt").read_text() + assert __file__ == wc.filename + with self.assertWarns(EncodingWarning) as wc: + root.joinpath("a.txt").open("r").close() + assert __file__ == wc.filename + + def test_open_write(self): + """ + If the zipfile is open for write, it should be possible to + write bytes or text to it. + """ + zf = zipfile.Path(zipfile.ZipFile(io.BytesIO(), mode='w')) + with zf.joinpath('file.bin').open('wb') as strm: + strm.write(b'binary contents') + with zf.joinpath('file.txt').open('w', encoding="utf-8") as strm: + strm.write('text file') + + def test_open_extant_directory(self): + """ + Attempting to open a directory raises IsADirectoryError. + """ + zf = zipfile.Path(add_dirs(build_alpharep_fixture())) + with self.assertRaises(IsADirectoryError): + zf.joinpath('b').open() + + @pass_alpharep + def test_open_binary_invalid_args(self, alpharep): + root = zipfile.Path(alpharep) + with self.assertRaises(ValueError): + root.joinpath('a.txt').open('rb', encoding='utf-8') + with self.assertRaises(ValueError): + root.joinpath('a.txt').open('rb', 'utf-8') + + def test_open_missing_directory(self): + """ + Attempting to open a missing directory raises FileNotFoundError. + """ + zf = zipfile.Path(add_dirs(build_alpharep_fixture())) + with self.assertRaises(FileNotFoundError): + zf.joinpath('z').open() + + @pass_alpharep + def test_read(self, alpharep): + root = zipfile.Path(alpharep) + a, b, g, j = root.iterdir() + assert a.read_text(encoding="utf-8") == "content of a" + # Also check positional encoding arg (gh-101144). + assert a.read_text("utf-8") == "content of a" + assert a.read_bytes() == b"content of a" + + @pass_alpharep + def test_joinpath(self, alpharep): + root = zipfile.Path(alpharep) + a = root.joinpath("a.txt") + assert a.is_file() + e = root.joinpath("b").joinpath("d").joinpath("e.txt") + assert e.read_text(encoding="utf-8") == "content of e" + + @pass_alpharep + def test_joinpath_multiple(self, alpharep): + root = zipfile.Path(alpharep) + e = root.joinpath("b", "d", "e.txt") + assert e.read_text(encoding="utf-8") == "content of e" + + @pass_alpharep + def test_traverse_truediv(self, alpharep): + root = zipfile.Path(alpharep) + a = root / "a.txt" + assert a.is_file() + e = root / "b" / "d" / "e.txt" + assert e.read_text(encoding="utf-8") == "content of e" + + @pass_alpharep + def test_pathlike_construction(self, alpharep): + """ + zipfile.Path should be constructable from a path-like object + """ + zipfile_ondisk = self.zipfile_ondisk(alpharep) + pathlike = FakePath(str(zipfile_ondisk)) + zipfile.Path(pathlike) + + @pass_alpharep + def test_traverse_pathlike(self, alpharep): + root = zipfile.Path(alpharep) + root / FakePath("a") + + @pass_alpharep + def test_parent(self, alpharep): + root = zipfile.Path(alpharep) + assert (root / 'a').parent.at == '' + assert (root / 'a' / 'b').parent.at == 'a/' + + @pass_alpharep + def test_dir_parent(self, alpharep): + root = zipfile.Path(alpharep) + assert (root / 'b').parent.at == '' + assert (root / 'b/').parent.at == '' + + @pass_alpharep + def test_missing_dir_parent(self, alpharep): + root = zipfile.Path(alpharep) + assert (root / 'missing dir/').parent.at == '' + + @pass_alpharep + def test_mutability(self, alpharep): + """ + If the underlying zipfile is changed, the Path object should + reflect that change. + """ + root = zipfile.Path(alpharep) + a, b, g, j = root.iterdir() + alpharep.writestr('foo.txt', 'foo') + alpharep.writestr('bar/baz.txt', 'baz') + assert any(child.name == 'foo.txt' for child in root.iterdir()) + assert (root / 'foo.txt').read_text(encoding="utf-8") == 'foo' + (baz,) = (root / 'bar').iterdir() + assert baz.read_text(encoding="utf-8") == 'baz' + + HUGE_ZIPFILE_NUM_ENTRIES = 2**13 + + def huge_zipfile(self): + """Create a read-only zipfile with a huge number of entries entries.""" + strm = io.BytesIO() + zf = zipfile.ZipFile(strm, "w") + for entry in map(str, range(self.HUGE_ZIPFILE_NUM_ENTRIES)): + zf.writestr(entry, entry) + zf.mode = 'r' + return zf + + def test_joinpath_constant_time(self): + """ + Ensure joinpath on items in zipfile is linear time. + """ + root = zipfile.Path(self.huge_zipfile()) + entries = jaraco.itertools.Counter(root.iterdir()) + for entry in entries: + entry.joinpath('suffix') + # Check the file iterated all items + assert entries.count == self.HUGE_ZIPFILE_NUM_ENTRIES + + @pass_alpharep + def test_read_does_not_close(self, alpharep): + alpharep = self.zipfile_ondisk(alpharep) + with zipfile.ZipFile(alpharep) as file: + for rep in range(2): + zipfile.Path(file, 'a.txt').read_text(encoding="utf-8") + + @pass_alpharep + def test_subclass(self, alpharep): + class Subclass(zipfile.Path): + pass + + root = Subclass(alpharep) + assert isinstance(root / 'b', Subclass) + + @pass_alpharep + def test_filename(self, alpharep): + root = zipfile.Path(alpharep) + assert root.filename == pathlib.Path('alpharep.zip') + + @pass_alpharep + def test_root_name(self, alpharep): + """ + The name of the root should be the name of the zipfile + """ + root = zipfile.Path(alpharep) + assert root.name == 'alpharep.zip' == root.filename.name + + @pass_alpharep + def test_suffix(self, alpharep): + """ + The suffix of the root should be the suffix of the zipfile. + The suffix of each nested file is the final component's last suffix, if any. + Includes the leading period, just like pathlib.Path. + """ + root = zipfile.Path(alpharep) + assert root.suffix == '.zip' == root.filename.suffix + + b = root / "b.txt" + assert b.suffix == ".txt" + + c = root / "c" / "filename.tar.gz" + assert c.suffix == ".gz" + + d = root / "d" + assert d.suffix == "" + + @pass_alpharep + def test_suffixes(self, alpharep): + """ + The suffix of the root should be the suffix of the zipfile. + The suffix of each nested file is the final component's last suffix, if any. + Includes the leading period, just like pathlib.Path. + """ + root = zipfile.Path(alpharep) + assert root.suffixes == ['.zip'] == root.filename.suffixes + + b = root / 'b.txt' + assert b.suffixes == ['.txt'] + + c = root / 'c' / 'filename.tar.gz' + assert c.suffixes == ['.tar', '.gz'] + + d = root / 'd' + assert d.suffixes == [] + + e = root / '.hgrc' + assert e.suffixes == [] + + @pass_alpharep + def test_suffix_no_filename(self, alpharep): + alpharep.filename = None + root = zipfile.Path(alpharep) + assert root.joinpath('example').suffix == "" + assert root.joinpath('example').suffixes == [] + + @pass_alpharep + def test_stem(self, alpharep): + """ + The final path component, without its suffix + """ + root = zipfile.Path(alpharep) + assert root.stem == 'alpharep' == root.filename.stem + + b = root / "b.txt" + assert b.stem == "b" + + c = root / "c" / "filename.tar.gz" + assert c.stem == "filename.tar" + + d = root / "d" + assert d.stem == "d" + + assert (root / ".gitignore").stem == ".gitignore" + + @pass_alpharep + def test_root_parent(self, alpharep): + root = zipfile.Path(alpharep) + assert root.parent == pathlib.Path('.') + root.root.filename = 'foo/bar.zip' + assert root.parent == pathlib.Path('foo') + + @pass_alpharep + def test_root_unnamed(self, alpharep): + """ + It is an error to attempt to get the name + or parent of an unnamed zipfile. + """ + alpharep.filename = None + root = zipfile.Path(alpharep) + with self.assertRaises(TypeError): + root.name + with self.assertRaises(TypeError): + root.parent + + # .name and .parent should still work on subs + sub = root / "b" + assert sub.name == "b" + assert sub.parent + + @pass_alpharep + def test_match_and_glob(self, alpharep): + root = zipfile.Path(alpharep) + assert not root.match("*.txt") + + assert list(root.glob("b/c.*")) == [zipfile.Path(alpharep, "b/c.txt")] + assert list(root.glob("b/*.txt")) == [ + zipfile.Path(alpharep, "b/c.txt"), + zipfile.Path(alpharep, "b/f.txt"), + ] + + @pass_alpharep + def test_glob_recursive(self, alpharep): + root = zipfile.Path(alpharep) + files = root.glob("**/*.txt") + assert all(each.match("*.txt") for each in files) + + assert list(root.glob("**/*.txt")) == list(root.rglob("*.txt")) + + @pass_alpharep + def test_glob_dirs(self, alpharep): + root = zipfile.Path(alpharep) + assert list(root.glob('b')) == [zipfile.Path(alpharep, "b/")] + assert list(root.glob('b*')) == [zipfile.Path(alpharep, "b/")] + + @pass_alpharep + def test_glob_subdir(self, alpharep): + root = zipfile.Path(alpharep) + assert list(root.glob('g/h')) == [zipfile.Path(alpharep, "g/h/")] + assert list(root.glob('g*/h*')) == [zipfile.Path(alpharep, "g/h/")] + + @pass_alpharep + def test_glob_subdirs(self, alpharep): + root = zipfile.Path(alpharep) + + assert list(root.glob("*/i.txt")) == [] + assert list(root.rglob("*/i.txt")) == [zipfile.Path(alpharep, "g/h/i.txt")] + + @pass_alpharep + def test_glob_does_not_overmatch_dot(self, alpharep): + root = zipfile.Path(alpharep) + + assert list(root.glob("*.xt")) == [] + + @pass_alpharep + def test_glob_single_char(self, alpharep): + root = zipfile.Path(alpharep) + + assert list(root.glob("a?txt")) == [zipfile.Path(alpharep, "a.txt")] + assert list(root.glob("a[.]txt")) == [zipfile.Path(alpharep, "a.txt")] + assert list(root.glob("a[?]txt")) == [] + + @pass_alpharep + def test_glob_chars(self, alpharep): + root = zipfile.Path(alpharep) + + assert list(root.glob("j/?.b[ai][nz]")) == [ + zipfile.Path(alpharep, "j/k.bin"), + zipfile.Path(alpharep, "j/l.baz"), + ] + + def test_glob_empty(self): + root = zipfile.Path(zipfile.ZipFile(io.BytesIO(), 'w')) + with self.assertRaises(ValueError): + root.glob('') + + @pass_alpharep + def test_eq_hash(self, alpharep): + root = zipfile.Path(alpharep) + assert root == zipfile.Path(alpharep) + + assert root != (root / "a.txt") + assert (root / "a.txt") == (root / "a.txt") + + root = zipfile.Path(alpharep) + assert root in {root} + + @pass_alpharep + def test_is_symlink(self, alpharep): + """ + See python/cpython#82102 for symlink support beyond this object. + """ + + root = zipfile.Path(alpharep) + assert not root.is_symlink() + + @pass_alpharep + def test_relative_to(self, alpharep): + root = zipfile.Path(alpharep) + relative = root.joinpath("b", "c.txt").relative_to(root / "b") + assert str(relative) == "c.txt" + + relative = root.joinpath("b", "d", "e.txt").relative_to(root / "b") + assert str(relative) == "d/e.txt" + + @pass_alpharep + def test_inheritance(self, alpharep): + cls = type('PathChild', (zipfile.Path,), {}) + file = cls(alpharep).joinpath('some dir').parent + assert isinstance(file, cls) + + @parameterize( + ['alpharep', 'path_type', 'subpath'], + itertools.product( + alpharep_generators, + [str, FakePath], + ['', 'b/'], + ), + ) + def test_pickle(self, alpharep, path_type, subpath): + zipfile_ondisk = path_type(str(self.zipfile_ondisk(alpharep))) + + saved_1 = pickle.dumps(zipfile.Path(zipfile_ondisk, at=subpath)) + restored_1 = pickle.loads(saved_1) + first, *rest = restored_1.iterdir() + assert first.read_text(encoding='utf-8').startswith('content of ') + + @pass_alpharep + def test_extract_orig_with_implied_dirs(self, alpharep): + """ + A zip file wrapped in a Path should extract even with implied dirs. + """ + source_path = self.zipfile_ondisk(alpharep) + zf = zipfile.ZipFile(source_path) + # wrap the zipfile for its side effect + zipfile.Path(zf) + zf.extractall(source_path.parent) + + @pass_alpharep + def test_getinfo_missing(self, alpharep): + """ + Validate behavior of getinfo on original zipfile after wrapping. + """ + zipfile.Path(alpharep) + with self.assertRaises(KeyError): + alpharep.getinfo('does-not-exist') + + def test_malformed_paths(self): + """ + Path should handle malformed paths gracefully. + + Paths with leading slashes are not visible. + + Paths with dots are treated like regular files. + """ + data = io.BytesIO() + zf = zipfile.ZipFile(data, "w") + zf.writestr("/one-slash.txt", b"content") + zf.writestr("//two-slash.txt", b"content") + zf.writestr("../parent.txt", b"content") + zf.filename = '' + root = zipfile.Path(zf) + assert list(map(str, root.iterdir())) == ['../'] + assert root.joinpath('..').joinpath('parent.txt').read_bytes() == b'content' + + def test_unsupported_names(self): + """ + Path segments with special characters are readable. + + On some platforms or file systems, characters like + ``:`` and ``?`` are not allowed, but they are valid + in the zip file. + """ + data = io.BytesIO() + zf = zipfile.ZipFile(data, "w") + zf.writestr("path?", b"content") + zf.writestr("V: NMS.flac", b"fLaC...") + zf.filename = '' + root = zipfile.Path(zf) + contents = root.iterdir() + assert next(contents).name == 'path?' + assert next(contents).name == 'V: NMS.flac' + assert root.joinpath('V: NMS.flac').read_bytes() == b"fLaC..." + + def test_backslash_not_separator(self): + """ + In a zip file, backslashes are not separators. + """ + data = io.BytesIO() + zf = zipfile.ZipFile(data, "w") + zf.writestr(DirtyZipInfo.for_name("foo\\bar", zf), b"content") + zf.filename = '' + root = zipfile.Path(zf) + (first,) = root.iterdir() + assert not first.is_dir() + assert first.name == 'foo\\bar' + + @pass_alpharep + def test_interface(self, alpharep): + from importlib.resources.abc import Traversable + + zf = zipfile.Path(alpharep) + assert isinstance(zf, Traversable) + + +class DirtyZipInfo(zipfile.ZipInfo): + """ + Bypass name sanitization. + """ + + def __init__(self, filename, *args, **kwargs): + super().__init__(filename, *args, **kwargs) + self.filename = filename + + @classmethod + def for_name(cls, name, archive): + """ + Construct the same way that ZipFile.writestr does. + + TODO: extract this functionality and re-use + """ + self = cls(filename=name, date_time=time.localtime(time.time())[:6]) + self.compress_type = archive.compression + self.compress_level = archive.compresslevel + if self.filename.endswith('/'): # pragma: no cover + self.external_attr = 0o40775 << 16 # drwxrwxr-x + self.external_attr |= 0x10 # MS-DOS directory flag + else: + self.external_attr = 0o600 << 16 # ?rw------- + return self diff --git a/cpython-tests/3.12/test_zipfile/_path/write-alpharep.py b/cpython-tests/3.12/test_zipfile/_path/write-alpharep.py new file mode 100644 index 0000000..48c09b5 --- /dev/null +++ b/cpython-tests/3.12/test_zipfile/_path/write-alpharep.py @@ -0,0 +1,4 @@ +from . import test_path + + +__name__ == '__main__' and test_path.build_alpharep_fixture().extractall('alpharep') diff --git a/cpython-tests/3.12/test_zipfile/test_core.py b/cpython-tests/3.12/test_zipfile/test_core.py new file mode 100644 index 0000000..cff8412 --- /dev/null +++ b/cpython-tests/3.12/test_zipfile/test_core.py @@ -0,0 +1,3530 @@ +import _pyio +import array +import contextlib +import importlib.util +import io +import itertools +import os +import posixpath +import struct +import subprocess +import sys +import time +import unittest +import unittest.mock as mock +import zipfile + + +from tempfile import TemporaryFile +from random import randint, random, randbytes + +import zipfile2 + +from test import archiver_tests +from test.support import script_helper +from test.support import ( + findfile, requires_zlib, requires_bz2, requires_lzma, + captured_stdout, captured_stderr, requires_subprocess +) +from test.support.os_helper import ( + TESTFN, unlink, rmtree, temp_dir, temp_cwd, fd_count, FakePath, + skip_unless_symlink +) + + +TESTFN2 = TESTFN + "2" +TESTFNDIR = TESTFN + "d" +FIXEDTEST_SIZE = 1000 +DATAFILES_DIR = 'zipfile_datafiles' + +SMALL_TEST_DATA = [('_ziptest1', '1q2w3e4r5t'), + ('ziptest2dir/_ziptest2', 'qawsedrftg'), + ('ziptest2dir/ziptest3dir/_ziptest3', 'azsxdcfvgb'), + ('ziptest2dir/ziptest3dir/ziptest4dir/_ziptest3', '6y7u8i9o0p')] + +def get_files(test): + yield TESTFN2 + with TemporaryFile() as f: + yield f + test.assertFalse(f.closed) + with io.BytesIO() as f: + yield f + test.assertFalse(f.closed) + +class AbstractTestsWithSourceFile: + @classmethod + def setUpClass(cls): + cls.line_gen = [bytes("Zipfile test line %d. random float: %f\n" % + (i, random()), "ascii") + for i in range(FIXEDTEST_SIZE)] + cls.data = b''.join(cls.line_gen) + + def setUp(self): + # Make a source file with some lines + with open(TESTFN, "wb") as fp: + fp.write(self.data) + + def make_test_archive(self, f, compression, compresslevel=None): + kwargs = {'compression': compression, 'compresslevel': compresslevel} + # Create the ZIP archive + with zipfile2.ZipFile(f, "w", **kwargs) as zipfp: + zipfp.write(TESTFN, "another.name") + zipfp.write(TESTFN, TESTFN) + zipfp.writestr("strfile", self.data) + with zipfp.open('written-open-w', mode='w') as f: + for line in self.line_gen: + f.write(line) + + def zip_test(self, f, compression, compresslevel=None): + self.make_test_archive(f, compression, compresslevel) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + self.assertEqual(zipfp.read(TESTFN), self.data) + self.assertEqual(zipfp.read("another.name"), self.data) + self.assertEqual(zipfp.read("strfile"), self.data) + + # Print the ZIP directory + fp = io.StringIO() + zipfp.printdir(file=fp) + directory = fp.getvalue() + lines = directory.splitlines() + self.assertEqual(len(lines), 5) # Number of files + header + + self.assertIn('File Name', lines[0]) + self.assertIn('Modified', lines[0]) + self.assertIn('Size', lines[0]) + + fn, date, time_, size = lines[1].split() + self.assertEqual(fn, 'another.name') + self.assertTrue(time.strptime(date, '%Y-%m-%d')) + self.assertTrue(time.strptime(time_, '%H:%M:%S')) + self.assertEqual(size, str(len(self.data))) + + # Check the namelist + names = zipfp.namelist() + self.assertEqual(len(names), 4) + self.assertIn(TESTFN, names) + self.assertIn("another.name", names) + self.assertIn("strfile", names) + self.assertIn("written-open-w", names) + + # Check infolist + infos = zipfp.infolist() + names = [i.filename for i in infos] + self.assertEqual(len(names), 4) + self.assertIn(TESTFN, names) + self.assertIn("another.name", names) + self.assertIn("strfile", names) + self.assertIn("written-open-w", names) + for i in infos: + self.assertEqual(i.file_size, len(self.data)) + + # check getinfo + for nm in (TESTFN, "another.name", "strfile", "written-open-w"): + info = zipfp.getinfo(nm) + self.assertEqual(info.filename, nm) + self.assertEqual(info.file_size, len(self.data)) + + # Check that testzip thinks the archive is ok + # (it returns None if all contents could be read properly) + self.assertIsNone(zipfp.testzip()) + + def test_basic(self): + for f in get_files(self): + self.zip_test(f, self.compression) + + def zip_open_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + zipdata1 = [] + with zipfp.open(TESTFN) as zipopen1: + while True: + read_data = zipopen1.read(256) + if not read_data: + break + zipdata1.append(read_data) + + zipdata2 = [] + with zipfp.open("another.name") as zipopen2: + while True: + read_data = zipopen2.read(256) + if not read_data: + break + zipdata2.append(read_data) + + self.assertEqual(b''.join(zipdata1), self.data) + self.assertEqual(b''.join(zipdata2), self.data) + + def test_open(self): + for f in get_files(self): + self.zip_open_test(f, self.compression) + + def test_open_with_pathlike(self): + path = FakePath(TESTFN2) + self.zip_open_test(path, self.compression) + with zipfile2.ZipFile(path, "r", self.compression) as zipfp: + self.assertIsInstance(zipfp.filename, str) + + def zip_random_open_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + zipdata1 = [] + with zipfp.open(TESTFN) as zipopen1: + while True: + read_data = zipopen1.read(randint(1, 1024)) + if not read_data: + break + zipdata1.append(read_data) + + self.assertEqual(b''.join(zipdata1), self.data) + + def test_random_open(self): + for f in get_files(self): + self.zip_random_open_test(f, self.compression) + + def zip_read1_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp, \ + zipfp.open(TESTFN) as zipopen: + zipdata = [] + while True: + read_data = zipopen.read1(-1) + if not read_data: + break + zipdata.append(read_data) + + self.assertEqual(b''.join(zipdata), self.data) + + def test_read1(self): + for f in get_files(self): + self.zip_read1_test(f, self.compression) + + def zip_read1_10_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp, \ + zipfp.open(TESTFN) as zipopen: + zipdata = [] + while True: + read_data = zipopen.read1(10) + self.assertLessEqual(len(read_data), 10) + if not read_data: + break + zipdata.append(read_data) + + self.assertEqual(b''.join(zipdata), self.data) + + def test_read1_10(self): + for f in get_files(self): + self.zip_read1_10_test(f, self.compression) + + def zip_readline_read_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp, \ + zipfp.open(TESTFN) as zipopen: + data = b'' + while True: + read = zipopen.readline() + if not read: + break + data += read + + read = zipopen.read(100) + if not read: + break + data += read + + self.assertEqual(data, self.data) + + def test_readline_read(self): + # Issue #7610: calls to readline() interleaved with calls to read(). + for f in get_files(self): + self.zip_readline_read_test(f, self.compression) + + def zip_readline_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp: + with zipfp.open(TESTFN) as zipopen: + for line in self.line_gen: + linedata = zipopen.readline() + self.assertEqual(linedata, line) + + def test_readline(self): + for f in get_files(self): + self.zip_readline_test(f, self.compression) + + def zip_readlines_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp: + with zipfp.open(TESTFN) as zipopen: + ziplines = zipopen.readlines() + for line, zipline in zip(self.line_gen, ziplines): + self.assertEqual(zipline, line) + + def test_readlines(self): + for f in get_files(self): + self.zip_readlines_test(f, self.compression) + + def zip_iterlines_test(self, f, compression): + self.make_test_archive(f, compression) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r") as zipfp: + with zipfp.open(TESTFN) as zipopen: + for line, zipline in zip(self.line_gen, zipopen): + self.assertEqual(zipline, line) + + def test_iterlines(self): + for f in get_files(self): + self.zip_iterlines_test(f, self.compression) + + def test_low_compression(self): + """Check for cases where compressed data is larger than original.""" + # Create the ZIP archive + with zipfile2.ZipFile(TESTFN2, "w", self.compression) as zipfp: + zipfp.writestr("strfile", '12') + + # Get an open object for strfile + with zipfile2.ZipFile(TESTFN2, "r", self.compression) as zipfp: + with zipfp.open("strfile") as openobj: + self.assertEqual(openobj.read(1), b'1') + self.assertEqual(openobj.read(1), b'2') + + def test_writestr_compression(self): + zipfp = zipfile2.ZipFile(TESTFN2, "w") + zipfp.writestr("b.txt", "hello world", compress_type=self.compression) + info = zipfp.getinfo('b.txt') + self.assertEqual(info.compress_type, self.compression) + + def test_writestr_compresslevel(self): + zipfp = zipfile2.ZipFile(TESTFN2, "w", compresslevel=1) + zipfp.writestr("a.txt", "hello world", compress_type=self.compression) + zipfp.writestr("b.txt", "hello world", compress_type=self.compression, + compresslevel=2) + + # Compression level follows the constructor. + a_info = zipfp.getinfo('a.txt') + self.assertEqual(a_info.compress_type, self.compression) + self.assertEqual(a_info._compresslevel, 1) + + # Compression level is overridden. + b_info = zipfp.getinfo('b.txt') + self.assertEqual(b_info.compress_type, self.compression) + self.assertEqual(b_info._compresslevel, 2) + + def test_read_return_size(self): + # Issue #9837: ZipExtFile.read() shouldn't return more bytes + # than requested. + for test_size in (1, 4095, 4096, 4097, 16384): + file_size = test_size + 1 + junk = randbytes(file_size) + with zipfile2.ZipFile(io.BytesIO(), "w", self.compression) as zipf: + zipf.writestr('foo', junk) + with zipf.open('foo', 'r') as fp: + buf = fp.read(test_size) + self.assertEqual(len(buf), test_size) + + def test_truncated_zipfile(self): + fp = io.BytesIO() + with zipfile2.ZipFile(fp, mode='w') as zipf: + zipf.writestr('strfile', self.data, compress_type=self.compression) + end_offset = fp.tell() + zipfiledata = fp.getvalue() + + fp = io.BytesIO(zipfiledata) + with zipfile2.ZipFile(fp) as zipf: + with zipf.open('strfile') as zipopen: + fp.truncate(end_offset - 20) + with self.assertRaises(EOFError): + zipopen.read() + + fp = io.BytesIO(zipfiledata) + with zipfile2.ZipFile(fp) as zipf: + with zipf.open('strfile') as zipopen: + fp.truncate(end_offset - 20) + with self.assertRaises(EOFError): + while zipopen.read(100): + pass + + fp = io.BytesIO(zipfiledata) + with zipfile2.ZipFile(fp) as zipf: + with zipf.open('strfile') as zipopen: + fp.truncate(end_offset - 20) + with self.assertRaises(EOFError): + while zipopen.read1(100): + pass + + def test_repr(self): + fname = 'file.name' + for f in get_files(self): + with zipfile2.ZipFile(f, 'w', self.compression) as zipfp: + zipfp.write(TESTFN, fname) + r = repr(zipfp) + self.assertIn("mode='w'", r) + + with zipfile2.ZipFile(f, 'r') as zipfp: + r = repr(zipfp) + if isinstance(f, str): + self.assertIn('filename=%r' % f, r) + else: + self.assertIn('file=%r' % f, r) + self.assertIn("mode='r'", r) + r = repr(zipfp.getinfo(fname)) + self.assertIn('filename=%r' % fname, r) + self.assertIn('filemode=', r) + self.assertIn('file_size=', r) + if self.compression != zipfile.ZIP_STORED: + self.assertIn('compress_type=', r) + self.assertIn('compress_size=', r) + with zipfp.open(fname) as zipopen: + r = repr(zipopen) + self.assertIn('name=%r' % fname, r) + self.assertIn("mode='r'", r) + if self.compression != zipfile.ZIP_STORED: + self.assertIn('compress_type=', r) + self.assertIn('[closed]', repr(zipopen)) + self.assertIn('[closed]', repr(zipfp)) + + def test_compresslevel_basic(self): + for f in get_files(self): + self.zip_test(f, self.compression, compresslevel=9) + + def test_per_file_compresslevel(self): + """Check that files within a Zip archive can have different + compression levels.""" + with zipfile2.ZipFile(TESTFN2, "w", compresslevel=1) as zipfp: + zipfp.write(TESTFN, 'compress_1') + zipfp.write(TESTFN, 'compress_9', compresslevel=9) + one_info = zipfp.getinfo('compress_1') + nine_info = zipfp.getinfo('compress_9') + self.assertEqual(one_info._compresslevel, 1) + self.assertEqual(nine_info._compresslevel, 9) + + def test_writing_errors(self): + class BrokenFile(io.BytesIO): + def write(self, data): + nonlocal count + if count is not None: + if count == stop: + raise OSError + count += 1 + super().write(data) + + stop = 0 + while True: + testfile = BrokenFile() + count = None + with zipfile2.ZipFile(testfile, 'w', self.compression) as zipfp: + with zipfp.open('file1', 'w') as f: + f.write(b'data1') + count = 0 + try: + with zipfp.open('file2', 'w') as f: + f.write(b'data2') + except OSError: + stop += 1 + else: + break + finally: + count = None + with zipfile2.ZipFile(io.BytesIO(testfile.getvalue())) as zipfp: + self.assertEqual(zipfp.namelist(), ['file1']) + self.assertEqual(zipfp.read('file1'), b'data1') + + with zipfile2.ZipFile(io.BytesIO(testfile.getvalue())) as zipfp: + self.assertEqual(zipfp.namelist(), ['file1', 'file2']) + self.assertEqual(zipfp.read('file1'), b'data1') + self.assertEqual(zipfp.read('file2'), b'data2') + + def test_zipextfile_attrs(self): + fname = "somefile.txt" + with zipfile2.ZipFile(TESTFN2, mode="w") as zipfp: + zipfp.writestr(fname, "bogus") + + with zipfile2.ZipFile(TESTFN2, mode="r") as zipfp: + with zipfp.open(fname) as fid: + self.assertEqual(fid.name, fname) + self.assertRaises(io.UnsupportedOperation, fid.fileno) + self.assertEqual(fid.mode, 'r') + self.assertIs(fid.readable(), True) + self.assertIs(fid.writable(), False) + self.assertIs(fid.seekable(), True) + self.assertIs(fid.closed, False) + self.assertIs(fid.closed, True) + self.assertEqual(fid.name, fname) + self.assertEqual(fid.mode, 'r') + self.assertRaises(io.UnsupportedOperation, fid.fileno) + self.assertRaises(ValueError, fid.readable) + self.assertIs(fid.writable(), False) + self.assertRaises(ValueError, fid.seekable) + + def tearDown(self): + unlink(TESTFN) + unlink(TESTFN2) + + +class StoredTestsWithSourceFile(AbstractTestsWithSourceFile, + unittest.TestCase): + compression = zipfile.ZIP_STORED + test_low_compression = None + + def zip_test_writestr_permissions(self, f, compression): + # Make sure that writestr and open(... mode='w') create files with + # mode 0600, when they are passed a name rather than a ZipInfo + # instance. + + self.make_test_archive(f, compression) + with zipfile2.ZipFile(f, "r") as zipfp: + zinfo = zipfp.getinfo('strfile') + self.assertEqual(zinfo.external_attr, 0o600 << 16) + + zinfo2 = zipfp.getinfo('written-open-w') + self.assertEqual(zinfo2.external_attr, 0o600 << 16) + + def test_writestr_permissions(self): + for f in get_files(self): + self.zip_test_writestr_permissions(f, zipfile.ZIP_STORED) + + def test_absolute_arcnames(self): + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, "/absolute") + + with zipfile2.ZipFile(TESTFN2, "r", zipfile.ZIP_STORED) as zipfp: + self.assertEqual(zipfp.namelist(), ["absolute"]) + + def test_append_to_zip_file(self): + """Test appending to an existing zipfile.""" + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + + with zipfile2.ZipFile(TESTFN2, "a", zipfile.ZIP_STORED) as zipfp: + zipfp.writestr("strfile", self.data) + self.assertEqual(zipfp.namelist(), [TESTFN, "strfile"]) + + def test_append_to_non_zip_file(self): + """Test appending to an existing file that is not a zipfile.""" + # NOTE: this test fails if len(d) < 22 because of the first + # line "fpin.seek(-22, 2)" in _EndRecData + data = b'I am not a ZipFile!'*10 + with open(TESTFN2, 'wb') as f: + f.write(data) + + with zipfile2.ZipFile(TESTFN2, "a", zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + + with open(TESTFN2, 'rb') as f: + f.seek(len(data)) + with zipfile2.ZipFile(f, "r") as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN]) + self.assertEqual(zipfp.read(TESTFN), self.data) + with open(TESTFN2, 'rb') as f: + self.assertEqual(f.read(len(data)), data) + zipfiledata = f.read() + with io.BytesIO(zipfiledata) as bio, zipfile2.ZipFile(bio) as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN]) + self.assertEqual(zipfp.read(TESTFN), self.data) + + def test_read_concatenated_zip_file(self): + with io.BytesIO() as bio: + with zipfile2.ZipFile(bio, 'w', zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + zipfiledata = bio.getvalue() + data = b'I am not a ZipFile!'*10 + with open(TESTFN2, 'wb') as f: + f.write(data) + f.write(zipfiledata) + + with zipfile2.ZipFile(TESTFN2) as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN]) + self.assertEqual(zipfp.read(TESTFN), self.data) + + def test_append_to_concatenated_zip_file(self): + with io.BytesIO() as bio: + with zipfile2.ZipFile(bio, 'w', zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + zipfiledata = bio.getvalue() + data = b'I am not a ZipFile!'*1000000 + with open(TESTFN2, 'wb') as f: + f.write(data) + f.write(zipfiledata) + + with zipfile2.ZipFile(TESTFN2, 'a') as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN]) + zipfp.writestr('strfile', self.data) + + with open(TESTFN2, 'rb') as f: + self.assertEqual(f.read(len(data)), data) + zipfiledata = f.read() + with io.BytesIO(zipfiledata) as bio, zipfile2.ZipFile(bio) as zipfp: + self.assertEqual(zipfp.namelist(), [TESTFN, 'strfile']) + self.assertEqual(zipfp.read(TESTFN), self.data) + self.assertEqual(zipfp.read('strfile'), self.data) + + def test_ignores_newline_at_end(self): + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED) as zipfp: + zipfp.write(TESTFN, TESTFN) + with open(TESTFN2, 'a', encoding='utf-8') as f: + f.write("\r\n\00\00\00") + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertIsInstance(zipfp, zipfile2.ZipFile) + + def test_ignores_stuff_appended_past_comments(self): + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED) as zipfp: + zipfp.comment = b"this is a comment" + zipfp.write(TESTFN, TESTFN) + with open(TESTFN2, 'a', encoding='utf-8') as f: + f.write("abcdef\r\n") + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertIsInstance(zipfp, zipfile2.ZipFile) + self.assertEqual(zipfp.comment, b"this is a comment") + + def test_write_default_name(self): + """Check that calling ZipFile.write without arcname specified + produces the expected result.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + zipfp.write(TESTFN) + with open(TESTFN, "rb") as f: + self.assertEqual(zipfp.read(TESTFN), f.read()) + + def test_io_on_closed_zipextfile(self): + fname = "somefile.txt" + with zipfile2.ZipFile(TESTFN2, mode="w", compression=self.compression) as zipfp: + zipfp.writestr(fname, "bogus") + + with zipfile2.ZipFile(TESTFN2, mode="r") as zipfp: + with zipfp.open(fname) as fid: + fid.close() + self.assertIs(fid.closed, True) + self.assertRaises(ValueError, fid.read) + self.assertRaises(ValueError, fid.seek, 0) + self.assertRaises(ValueError, fid.tell) + + def test_write_to_readonly(self): + """Check that trying to call write() on a readonly ZipFile object + raises a ValueError.""" + with zipfile2.ZipFile(TESTFN2, mode="w") as zipfp: + zipfp.writestr("somefile.txt", "bogus") + + with zipfile2.ZipFile(TESTFN2, mode="r") as zipfp: + self.assertRaises(ValueError, zipfp.write, TESTFN) + + with zipfile2.ZipFile(TESTFN2, mode="r") as zipfp: + with self.assertRaises(ValueError): + zipfp.open(TESTFN, mode='w') + + def test_add_file_before_1980(self): + # Set atime and mtime to 1970-01-01 + os.utime(TESTFN, (0, 0)) + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + self.assertRaises(ValueError, zipfp.write, TESTFN) + + with zipfile2.ZipFile(TESTFN2, "w", strict_timestamps=False) as zipfp: + zipfp.write(TESTFN) + zinfo = zipfp.getinfo(TESTFN) + self.assertEqual(zinfo.date_time, (1980, 1, 1, 0, 0, 0)) + + def test_add_file_after_2107(self): + # Set atime and mtime to 2108-12-30 + ts = 4386268800 + try: + time.localtime(ts) + except OverflowError: + self.skipTest(f'time.localtime({ts}) raises OverflowError') + try: + os.utime(TESTFN, (ts, ts)) + except OverflowError: + self.skipTest('Host fs cannot set timestamp to required value.') + + mtime_ns = os.stat(TESTFN).st_mtime_ns + if mtime_ns != (4386268800 * 10**9): + # XFS filesystem is limited to 32-bit timestamp, but the syscall + # didn't fail. Moreover, there is a VFS bug which returns + # a cached timestamp which is different than the value on disk. + # + # Test st_mtime_ns rather than st_mtime to avoid rounding issues. + # + # https://bugzilla.redhat.com/show_bug.cgi?id=1795576 + # https://bugs.python.org/issue39460#msg360952 + self.skipTest(f"Linux VFS/XFS kernel bug detected: {mtime_ns=}") + + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + self.assertRaises(struct.error, zipfp.write, TESTFN) + + with zipfile2.ZipFile(TESTFN2, "w", strict_timestamps=False) as zipfp: + zipfp.write(TESTFN) + zinfo = zipfp.getinfo(TESTFN) + self.assertEqual(zinfo.date_time, (2107, 12, 31, 23, 59, 59)) + + +@requires_zlib() +class DeflateTestsWithSourceFile(AbstractTestsWithSourceFile, + unittest.TestCase): + compression = zipfile.ZIP_DEFLATED + + def test_per_file_compression(self): + """Check that files within a Zip archive can have different + compression options.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + zipfp.write(TESTFN, 'storeme', zipfile.ZIP_STORED) + zipfp.write(TESTFN, 'deflateme', zipfile.ZIP_DEFLATED) + sinfo = zipfp.getinfo('storeme') + dinfo = zipfp.getinfo('deflateme') + self.assertEqual(sinfo.compress_type, zipfile.ZIP_STORED) + self.assertEqual(dinfo.compress_type, zipfile.ZIP_DEFLATED) + +@requires_bz2() +class Bzip2TestsWithSourceFile(AbstractTestsWithSourceFile, + unittest.TestCase): + compression = zipfile.ZIP_BZIP2 + +@requires_lzma() +class LzmaTestsWithSourceFile(AbstractTestsWithSourceFile, + unittest.TestCase): + compression = zipfile.ZIP_LZMA + + +class AbstractTestZip64InSmallFiles: + # These tests test the ZIP64 functionality without using large files, + # see test_zipfile64 for proper tests. + + @classmethod + def setUpClass(cls): + line_gen = (bytes("Test of zipfile line %d." % i, "ascii") + for i in range(0, FIXEDTEST_SIZE)) + cls.data = b'\n'.join(line_gen) + + def setUp(self): + self._limit = zipfile.ZIP64_LIMIT + self._filecount_limit = zipfile.ZIP_FILECOUNT_LIMIT + zipfile.ZIP64_LIMIT = 1000 + zipfile.ZIP_FILECOUNT_LIMIT = 9 + + # Make a source file with some lines + with open(TESTFN, "wb") as fp: + fp.write(self.data) + + def zip_test(self, f, compression): + # Create the ZIP archive + with zipfile2.ZipFile(f, "w", compression, allowZip64=True) as zipfp: + zipfp.write(TESTFN, "another.name") + zipfp.write(TESTFN, TESTFN) + zipfp.writestr("strfile", self.data) + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + self.assertEqual(zipfp.read(TESTFN), self.data) + self.assertEqual(zipfp.read("another.name"), self.data) + self.assertEqual(zipfp.read("strfile"), self.data) + + # Print the ZIP directory + fp = io.StringIO() + zipfp.printdir(fp) + + directory = fp.getvalue() + lines = directory.splitlines() + self.assertEqual(len(lines), 4) # Number of files + header + + self.assertIn('File Name', lines[0]) + self.assertIn('Modified', lines[0]) + self.assertIn('Size', lines[0]) + + fn, date, time_, size = lines[1].split() + self.assertEqual(fn, 'another.name') + self.assertTrue(time.strptime(date, '%Y-%m-%d')) + self.assertTrue(time.strptime(time_, '%H:%M:%S')) + self.assertEqual(size, str(len(self.data))) + + # Check the namelist + names = zipfp.namelist() + self.assertEqual(len(names), 3) + self.assertIn(TESTFN, names) + self.assertIn("another.name", names) + self.assertIn("strfile", names) + + # Check infolist + infos = zipfp.infolist() + names = [i.filename for i in infos] + self.assertEqual(len(names), 3) + self.assertIn(TESTFN, names) + self.assertIn("another.name", names) + self.assertIn("strfile", names) + for i in infos: + self.assertEqual(i.file_size, len(self.data)) + + # check getinfo + for nm in (TESTFN, "another.name", "strfile"): + info = zipfp.getinfo(nm) + self.assertEqual(info.filename, nm) + self.assertEqual(info.file_size, len(self.data)) + + # Check that testzip thinks the archive is valid + self.assertIsNone(zipfp.testzip()) + + def test_basic(self): + for f in get_files(self): + self.zip_test(f, self.compression) + + def test_too_many_files(self): + # This test checks that more than 64k files can be added to an archive, + # and that the resulting archive can be read properly by ZipFile + zipf = zipfile2.ZipFile(TESTFN, "w", self.compression, + allowZip64=True) + zipf.debug = 100 + numfiles = 15 + for i in range(numfiles): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles) + zipf.close() + + zipf2 = zipfile2.ZipFile(TESTFN, "r", self.compression) + self.assertEqual(len(zipf2.namelist()), numfiles) + for i in range(numfiles): + content = zipf2.read("foo%08d" % i).decode('ascii') + self.assertEqual(content, "%d" % (i**3 % 57)) + zipf2.close() + + def test_too_many_files_append(self): + zipf = zipfile2.ZipFile(TESTFN, "w", self.compression, + allowZip64=False) + zipf.debug = 100 + numfiles = 9 + for i in range(numfiles): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles) + with self.assertRaises(zipfile.LargeZipFile): + zipf.writestr("foo%08d" % numfiles, b'') + self.assertEqual(len(zipf.namelist()), numfiles) + zipf.close() + + zipf = zipfile2.ZipFile(TESTFN, "a", self.compression, + allowZip64=False) + zipf.debug = 100 + self.assertEqual(len(zipf.namelist()), numfiles) + with self.assertRaises(zipfile.LargeZipFile): + zipf.writestr("foo%08d" % numfiles, b'') + self.assertEqual(len(zipf.namelist()), numfiles) + zipf.close() + + zipf = zipfile2.ZipFile(TESTFN, "a", self.compression, + allowZip64=True) + zipf.debug = 100 + self.assertEqual(len(zipf.namelist()), numfiles) + numfiles2 = 15 + for i in range(numfiles, numfiles2): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles2) + zipf.close() + + zipf2 = zipfile2.ZipFile(TESTFN, "r", self.compression) + self.assertEqual(len(zipf2.namelist()), numfiles2) + for i in range(numfiles2): + content = zipf2.read("foo%08d" % i).decode('ascii') + self.assertEqual(content, "%d" % (i**3 % 57)) + zipf2.close() + + def tearDown(self): + zipfile.ZIP64_LIMIT = self._limit + zipfile.ZIP_FILECOUNT_LIMIT = self._filecount_limit + unlink(TESTFN) + unlink(TESTFN2) + + +class StoredTestZip64InSmallFiles(AbstractTestZip64InSmallFiles, + unittest.TestCase): + compression = zipfile.ZIP_STORED + + def large_file_exception_test(self, f, compression): + with zipfile2.ZipFile(f, "w", compression, allowZip64=False) as zipfp: + self.assertRaises(zipfile.LargeZipFile, + zipfp.write, TESTFN, "another.name") + + def large_file_exception_test2(self, f, compression): + with zipfile2.ZipFile(f, "w", compression, allowZip64=False) as zipfp: + self.assertRaises(zipfile.LargeZipFile, + zipfp.writestr, "another.name", self.data) + + def test_large_file_exception(self): + for f in get_files(self): + self.large_file_exception_test(f, zipfile.ZIP_STORED) + self.large_file_exception_test2(f, zipfile.ZIP_STORED) + + def test_absolute_arcnames(self): + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED, + allowZip64=True) as zipfp: + zipfp.write(TESTFN, "/absolute") + + with zipfile2.ZipFile(TESTFN2, "r", zipfile.ZIP_STORED) as zipfp: + self.assertEqual(zipfp.namelist(), ["absolute"]) + + def test_append(self): + # Test that appending to the Zip64 archive doesn't change + # extra fields of existing entries. + with zipfile2.ZipFile(TESTFN2, "w", allowZip64=True) as zipfp: + zipfp.writestr("strfile", self.data) + with zipfile2.ZipFile(TESTFN2, "r", allowZip64=True) as zipfp: + zinfo = zipfp.getinfo("strfile") + extra = zinfo.extra + with zipfile2.ZipFile(TESTFN2, "a", allowZip64=True) as zipfp: + zipfp.writestr("strfile2", self.data) + with zipfile2.ZipFile(TESTFN2, "r", allowZip64=True) as zipfp: + zinfo = zipfp.getinfo("strfile") + self.assertEqual(zinfo.extra, extra) + + def make_zip64_file( + self, file_size_64_set=False, file_size_extra=False, + compress_size_64_set=False, compress_size_extra=False, + header_offset_64_set=False, header_offset_extra=False, + ): + """Generate bytes sequence for a zip with (incomplete) zip64 data. + + The actual values (not the zip 64 0xffffffff values) stored in the file + are: + file_size: 8 + compress_size: 8 + header_offset: 0 + """ + actual_size = 8 + actual_header_offset = 0 + local_zip64_fields = [] + central_zip64_fields = [] + + file_size = actual_size + if file_size_64_set: + file_size = 0xffffffff + if file_size_extra: + local_zip64_fields.append(actual_size) + central_zip64_fields.append(actual_size) + file_size = struct.pack("= ZIP64_VERSION + # - The compressed and uncompressed size in the file headers are both + # 0xFFFFFFFF (ie. point to zip64 record) + # - The zip64 record is provided and has the correct sizes in it + # Other aspects of the zip are checked as well, but verifying the above is the main goal. + # Because this is hard to verify by parsing the data as a zip, the raw + # bytes are checked to ensure that they line up with the zip spec. + # The spec for this can be found at: https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT + # The relevent sections for this test are: + # - 4.3.7 for local file header + # - 4.5.3 for zip64 extra field + + data = io.BytesIO() + with zipfile2.ZipFile(data, mode="w", allowZip64=True) as zf: + with zf.open("text.txt", mode="w", force_zip64=True) as zi: + zi.write(b"_") + + zipdata = data.getvalue() + + # pull out and check zip information + ( + header, vers, os, flags, comp, csize, usize, fn_len, + ex_total_len, filename, ex_id, ex_len, ex_usize, ex_csize, cd_sig + ) = struct.unpack("<4sBBHH8xIIHH8shhQQx4s", zipdata[:63]) + + self.assertEqual(header, b"PK\x03\x04") # local file header + self.assertGreaterEqual(vers, zipfile.ZIP64_VERSION) # requires zip64 to extract + self.assertEqual(os, 0) # compatible with MS-DOS + self.assertEqual(flags, 0) # no flags + self.assertEqual(comp, 0) # compression method = stored + self.assertEqual(csize, 0xFFFFFFFF) # sizes are in zip64 extra + self.assertEqual(usize, 0xFFFFFFFF) + self.assertEqual(fn_len, 8) # filename len + self.assertEqual(ex_total_len, 20) # size of extra records + self.assertEqual(ex_id, 1) # Zip64 extra record + self.assertEqual(ex_len, 16) # 16 bytes of data + self.assertEqual(ex_usize, 1) # uncompressed size + self.assertEqual(ex_csize, 1) # compressed size + self.assertEqual(cd_sig, b"PK\x01\x02") # ensure the central directory header is next + + z = zipfile2.ZipFile(io.BytesIO(zipdata)) + zinfos = z.infolist() + self.assertEqual(len(zinfos), 1) + self.assertGreaterEqual(zinfos[0].extract_version, zipfile.ZIP64_VERSION) # requires zip64 to extract + + def test_unseekable_zip_unknown_filesize(self): + """Test that creating a zip with/without seeking will raise a RuntimeError if zip64 was required but not used""" + + def make_zip(fp): + with zipfile2.ZipFile(fp, mode="w", allowZip64=True) as zf: + with zf.open("text.txt", mode="w", force_zip64=False) as zi: + zi.write(b"_" * (zipfile.ZIP64_LIMIT + 1)) + + self.assertRaises(RuntimeError, make_zip, io.BytesIO()) + self.assertRaises(RuntimeError, make_zip, Unseekable(io.BytesIO())) + + def test_zip64_required_not_allowed_fail(self): + """Test that trying to add a large file to a zip that doesn't allow zip64 extensions fails on add""" + def make_zip(fp): + with zipfile2.ZipFile(fp, mode="w", allowZip64=False) as zf: + # pretend zipfile.ZipInfo.from_file was used to get the name and filesize + info = zipfile.ZipInfo("text.txt") + info.file_size = zipfile.ZIP64_LIMIT + 1 + zf.open(info, mode="w") + + self.assertRaises(zipfile.LargeZipFile, make_zip, io.BytesIO()) + self.assertRaises(zipfile.LargeZipFile, make_zip, Unseekable(io.BytesIO())) + + def test_unseekable_zip_known_filesize(self): + """Test that creating a zip without seeking will use zip64 extensions if the file size is provided up-front""" + + # This test ensures that the zip will use a zip64 data descriptor (same + # as a regular data descriptor except the sizes are 8 bytes instead of + # 4) record to communicate the size of a file if the zip is being + # written to an unseekable stream. + # Because this sort of thing is hard to verify by parsing the data back + # in as a zip, this test looks at the raw bytes created to ensure that + # the correct data has been generated. + # The spec for this can be found at: https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT + # The relevant sections for this test are: + # - 4.3.7 for local file header + # - 4.3.9 for the data descriptor + # - 4.5.3 for zip64 extra field + + file_size = zipfile.ZIP64_LIMIT + 1 + + def make_zip(fp): + with zipfile2.ZipFile(fp, mode="w", allowZip64=True) as zf: + # pretend zipfile.ZipInfo.from_file was used to get the name and filesize + info = zipfile.ZipInfo("text.txt") + info.file_size = file_size + with zf.open(info, mode="w", force_zip64=False) as zi: + zi.write(b"_" * file_size) + return fp + + # check seekable file information + seekable_data = make_zip(io.BytesIO()).getvalue() + ( + header, vers, os, flags, comp, csize, usize, fn_len, + ex_total_len, filename, ex_id, ex_len, ex_usize, ex_csize, + cd_sig + ) = struct.unpack("<4sBBHH8xIIHH8shhQQ{}x4s".format(file_size), seekable_data[:62 + file_size]) + + self.assertEqual(header, b"PK\x03\x04") # local file header + self.assertGreaterEqual(vers, zipfile.ZIP64_VERSION) # requires zip64 to extract + self.assertEqual(os, 0) # compatible with MS-DOS + self.assertEqual(flags, 0) # no flags set + self.assertEqual(comp, 0) # compression method = stored + self.assertEqual(csize, 0xFFFFFFFF) # sizes are in zip64 extra + self.assertEqual(usize, 0xFFFFFFFF) + self.assertEqual(fn_len, 8) # filename len + self.assertEqual(ex_total_len, 20) # size of extra records + self.assertEqual(ex_id, 1) # Zip64 extra record + self.assertEqual(ex_len, 16) # 16 bytes of data + self.assertEqual(ex_usize, file_size) # uncompressed size + self.assertEqual(ex_csize, file_size) # compressed size + self.assertEqual(cd_sig, b"PK\x01\x02") # ensure the central directory header is next + + # check unseekable file information + unseekable_data = make_zip(Unseekable(io.BytesIO())).fp.getvalue() + ( + header, vers, os, flags, comp, csize, usize, fn_len, + ex_total_len, filename, ex_id, ex_len, ex_usize, ex_csize, + dd_header, dd_usize, dd_csize, cd_sig + ) = struct.unpack("<4sBBHH8xIIHH8shhQQ{}x4s4xQQ4s".format(file_size), unseekable_data[:86 + file_size]) + + self.assertEqual(header, b"PK\x03\x04") # local file header + self.assertGreaterEqual(vers, zipfile.ZIP64_VERSION) # requires zip64 to extract + self.assertEqual(os, 0) # compatible with MS-DOS + self.assertEqual("{:b}".format(flags), "1000") # streaming flag set + self.assertEqual(comp, 0) # compression method = stored + self.assertEqual(csize, 0xFFFFFFFF) # sizes are in zip64 extra + self.assertEqual(usize, 0xFFFFFFFF) + self.assertEqual(fn_len, 8) # filename len + self.assertEqual(ex_total_len, 20) # size of extra records + self.assertEqual(ex_id, 1) # Zip64 extra record + self.assertEqual(ex_len, 16) # 16 bytes of data + self.assertEqual(ex_usize, 0) # uncompressed size - 0 to defer to data descriptor + self.assertEqual(ex_csize, 0) # compressed size - 0 to defer to data descriptor + self.assertEqual(dd_header, b"PK\07\x08") # data descriptor + self.assertEqual(dd_usize, file_size) # file size (8 bytes because zip64) + self.assertEqual(dd_csize, file_size) # compressed size (8 bytes because zip64) + self.assertEqual(cd_sig, b"PK\x01\x02") # ensure the central directory header is next + + +@requires_zlib() +class DeflateTestZip64InSmallFiles(AbstractTestZip64InSmallFiles, + unittest.TestCase): + compression = zipfile.ZIP_DEFLATED + +@requires_bz2() +class Bzip2TestZip64InSmallFiles(AbstractTestZip64InSmallFiles, + unittest.TestCase): + compression = zipfile.ZIP_BZIP2 + +@requires_lzma() +class LzmaTestZip64InSmallFiles(AbstractTestZip64InSmallFiles, + unittest.TestCase): + compression = zipfile.ZIP_LZMA + + +class AbstractWriterTests: + + def tearDown(self): + unlink(TESTFN2) + + def test_close_after_close(self): + data = b'content' + with zipfile2.ZipFile(TESTFN2, "w", self.compression) as zipf: + w = zipf.open('test', 'w') + w.write(data) + w.close() + self.assertTrue(w.closed) + w.close() + self.assertTrue(w.closed) + self.assertEqual(zipf.read('test'), data) + + def test_write_after_close(self): + data = b'content' + with zipfile2.ZipFile(TESTFN2, "w", self.compression) as zipf: + w = zipf.open('test', 'w') + w.write(data) + w.close() + self.assertTrue(w.closed) + self.assertRaises(ValueError, w.write, b'') + self.assertEqual(zipf.read('test'), data) + + def test_issue44439(self): + q = array.array('Q', [1, 2, 3, 4, 5]) + LENGTH = len(q) * q.itemsize + with zipfile2.ZipFile(io.BytesIO(), 'w', self.compression) as zip: + with zip.open('data', 'w') as data: + self.assertEqual(data.write(q), LENGTH) + self.assertEqual(zip.getinfo('data').file_size, LENGTH) + + def test_zipwritefile_attrs(self): + fname = "somefile.txt" + with zipfile2.ZipFile(TESTFN2, mode="w", compression=self.compression) as zipfp: + with zipfp.open(fname, 'w') as fid: + self.assertRaises(io.UnsupportedOperation, fid.fileno) + self.assertIs(fid.readable(), False) + self.assertIs(fid.writable(), True) + self.assertIs(fid.seekable(), False) + self.assertIs(fid.closed, False) + self.assertIs(fid.closed, True) + self.assertRaises(io.UnsupportedOperation, fid.fileno) + self.assertIs(fid.readable(), False) + self.assertIs(fid.writable(), True) + self.assertIs(fid.seekable(), False) + +class StoredWriterTests(AbstractWriterTests, unittest.TestCase): + compression = zipfile.ZIP_STORED + +@requires_zlib() +class DeflateWriterTests(AbstractWriterTests, unittest.TestCase): + compression = zipfile.ZIP_DEFLATED + +@requires_bz2() +class Bzip2WriterTests(AbstractWriterTests, unittest.TestCase): + compression = zipfile.ZIP_BZIP2 + +@requires_lzma() +class LzmaWriterTests(AbstractWriterTests, unittest.TestCase): + compression = zipfile.ZIP_LZMA + +@unittest.skip('Zipfile2 does not support PyZipFile') +class PyZipFileTests(unittest.TestCase): + def assertCompiledIn(self, name, namelist): + if name + 'o' not in namelist: + self.assertIn(name + 'c', namelist) + + def requiresWriteAccess(self, path): + # effective_ids unavailable on windows + if not os.access(path, os.W_OK, + effective_ids=os.access in os.supports_effective_ids): + self.skipTest('requires write access to the installed location') + filename = os.path.join(path, 'test_zipfile.try') + try: + fd = os.open(filename, os.O_WRONLY | os.O_CREAT) + os.close(fd) + except Exception: + self.skipTest('requires write access to the installed location') + unlink(filename) + + def test_write_pyfile(self): + self.requiresWriteAccess(os.path.dirname(__file__)) + with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp: + fn = __file__ + if fn.endswith('.pyc'): + path_split = fn.split(os.sep) + if os.altsep is not None: + path_split.extend(fn.split(os.altsep)) + if '__pycache__' in path_split: + fn = importlib.util.source_from_cache(fn) + else: + fn = fn[:-1] + + zipfp.writepy(fn) + + bn = os.path.basename(fn) + self.assertNotIn(bn, zipfp.namelist()) + self.assertCompiledIn(bn, zipfp.namelist()) + + with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp: + fn = __file__ + if fn.endswith('.pyc'): + fn = fn[:-1] + + zipfp.writepy(fn, "testpackage") + + bn = "%s/%s" % ("testpackage", os.path.basename(fn)) + self.assertNotIn(bn, zipfp.namelist()) + self.assertCompiledIn(bn, zipfp.namelist()) + + def test_write_python_package(self): + import email + packagedir = os.path.dirname(email.__file__) + self.requiresWriteAccess(packagedir) + + with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp: + zipfp.writepy(packagedir) + + # Check for a couple of modules at different levels of the + # hierarchy + names = zipfp.namelist() + self.assertCompiledIn('email/__init__.py', names) + self.assertCompiledIn('email/mime/text.py', names) + + def test_write_filtered_python_package(self): + import test + packagedir = os.path.dirname(test.__file__) + self.requiresWriteAccess(packagedir) + + with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp: + + # first make sure that the test folder gives error messages + # (on the badsyntax_... files) + with captured_stdout() as reportSIO: + zipfp.writepy(packagedir) + reportStr = reportSIO.getvalue() + self.assertTrue('SyntaxError' in reportStr) + + # then check that the filter works on the whole package + with captured_stdout() as reportSIO: + zipfp.writepy(packagedir, filterfunc=lambda whatever: False) + reportStr = reportSIO.getvalue() + self.assertTrue('SyntaxError' not in reportStr) + + # then check that the filter works on individual files + def filter(path): + return not os.path.basename(path).startswith("bad") + with captured_stdout() as reportSIO, self.assertWarns(UserWarning): + zipfp.writepy(packagedir, filterfunc=filter) + reportStr = reportSIO.getvalue() + if reportStr: + print(reportStr) + self.assertTrue('SyntaxError' not in reportStr) + + def test_write_with_optimization(self): + import email + packagedir = os.path.dirname(email.__file__) + self.requiresWriteAccess(packagedir) + optlevel = 1 if __debug__ else 0 + ext = '.pyc' + + with TemporaryFile() as t, \ + zipfile.PyZipFile(t, "w", optimize=optlevel) as zipfp: + zipfp.writepy(packagedir) + + names = zipfp.namelist() + self.assertIn('email/__init__' + ext, names) + self.assertIn('email/mime/text' + ext, names) + + def test_write_python_directory(self): + os.mkdir(TESTFN2) + try: + with open(os.path.join(TESTFN2, "mod1.py"), "w", encoding='utf-8') as fp: + fp.write("print(42)\n") + + with open(os.path.join(TESTFN2, "mod2.py"), "w", encoding='utf-8') as fp: + fp.write("print(42 * 42)\n") + + with open(os.path.join(TESTFN2, "mod2.txt"), "w", encoding='utf-8') as fp: + fp.write("bla bla bla\n") + + with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp: + zipfp.writepy(TESTFN2) + + names = zipfp.namelist() + self.assertCompiledIn('mod1.py', names) + self.assertCompiledIn('mod2.py', names) + self.assertNotIn('mod2.txt', names) + + finally: + rmtree(TESTFN2) + + def test_write_python_directory_filtered(self): + os.mkdir(TESTFN2) + try: + with open(os.path.join(TESTFN2, "mod1.py"), "w", encoding='utf-8') as fp: + fp.write("print(42)\n") + + with open(os.path.join(TESTFN2, "mod2.py"), "w", encoding='utf-8') as fp: + fp.write("print(42 * 42)\n") + + with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp: + zipfp.writepy(TESTFN2, filterfunc=lambda fn: + not fn.endswith('mod2.py')) + + names = zipfp.namelist() + self.assertCompiledIn('mod1.py', names) + self.assertNotIn('mod2.py', names) + + finally: + rmtree(TESTFN2) + + def test_write_non_pyfile(self): + with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp: + with open(TESTFN, 'w', encoding='utf-8') as f: + f.write('most definitely not a python file') + self.assertRaises(RuntimeError, zipfp.writepy, TESTFN) + unlink(TESTFN) + + def test_write_pyfile_bad_syntax(self): + os.mkdir(TESTFN2) + try: + with open(os.path.join(TESTFN2, "mod1.py"), "w", encoding='utf-8') as fp: + fp.write("Bad syntax in python file\n") + + with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp: + # syntax errors are printed to stdout + with captured_stdout() as s: + zipfp.writepy(os.path.join(TESTFN2, "mod1.py")) + + self.assertIn("SyntaxError", s.getvalue()) + + # as it will not have compiled the python file, it will + # include the .py file not .pyc + names = zipfp.namelist() + self.assertIn('mod1.py', names) + self.assertNotIn('mod1.pyc', names) + + finally: + rmtree(TESTFN2) + + def test_write_pathlike(self): + os.mkdir(TESTFN2) + try: + with open(os.path.join(TESTFN2, "mod1.py"), "w", encoding='utf-8') as fp: + fp.write("print(42)\n") + + with TemporaryFile() as t, zipfile.PyZipFile(t, "w") as zipfp: + zipfp.writepy(FakePath(os.path.join(TESTFN2, "mod1.py"))) + names = zipfp.namelist() + self.assertCompiledIn('mod1.py', names) + finally: + rmtree(TESTFN2) + + +class ExtractTests(unittest.TestCase): + + def make_test_file(self): + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED) as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + + def test_extract(self): + with temp_cwd(): + self.make_test_file() + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + writtenfile = zipfp.extract(fpath) + + # make sure it was written to the right place + correctfile = os.path.join(os.getcwd(), fpath) + correctfile = os.path.normpath(correctfile) + + self.assertEqual(writtenfile, correctfile) + + # make sure correct data is in correct file + with open(writtenfile, "rb") as f: + self.assertEqual(fdata.encode(), f.read()) + + unlink(writtenfile) + + def _test_extract_with_target(self, target): + self.make_test_file() + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + writtenfile = zipfp.extract(fpath, target) + + # make sure it was written to the right place + correctfile = os.path.join(target, fpath) + correctfile = os.path.normpath(correctfile) + self.assertTrue(os.path.samefile(writtenfile, correctfile), (writtenfile, target)) + + # make sure correct data is in correct file + with open(writtenfile, "rb") as f: + self.assertEqual(fdata.encode(), f.read()) + + unlink(writtenfile) + + unlink(TESTFN2) + + def test_extract_with_target(self): + with temp_dir() as extdir: + self._test_extract_with_target(extdir) + + def test_extract_with_target_pathlike(self): + with temp_dir() as extdir: + self._test_extract_with_target(FakePath(extdir)) + + def test_extract_all(self): + with temp_cwd(): + self.make_test_file() + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + zipfp.extractall() + for fpath, fdata in SMALL_TEST_DATA: + outfile = os.path.join(os.getcwd(), fpath) + + with open(outfile, "rb") as f: + self.assertEqual(fdata.encode(), f.read()) + + unlink(outfile) + + def _test_extract_all_with_target(self, target): + self.make_test_file() + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + zipfp.extractall(target) + for fpath, fdata in SMALL_TEST_DATA: + outfile = os.path.join(target, fpath) + + with open(outfile, "rb") as f: + self.assertEqual(fdata.encode(), f.read()) + + unlink(outfile) + + unlink(TESTFN2) + + def test_extract_all_with_target(self): + with temp_dir() as extdir: + self._test_extract_all_with_target(extdir) + + def test_extract_all_with_target_pathlike(self): + with temp_dir() as extdir: + self._test_extract_all_with_target(FakePath(extdir)) + + def check_file(self, filename, content): + self.assertTrue(os.path.isfile(filename)) + with open(filename, 'rb') as f: + self.assertEqual(f.read(), content) + + def test_sanitize_windows_name(self): + san = zipfile2.ZipFile._sanitize_windows_name + # Passing pathsep in allows this test to work regardless of platform. + self.assertEqual(san(r',,?,C:,foo,bar/z', ','), r'_,C_,foo,bar/z') + self.assertEqual(san(r'a\b,ce|f"g?h*i', ','), r'a\b,c_d_e_f_g_h_i') + self.assertEqual(san('../../foo../../ba..r', '/'), r'foo/ba..r') + self.assertEqual(san(' / /foo / /ba r', '/'), r'foo/ba r') + self.assertEqual(san(' . /. /foo ./ . /. ./ba .r', '/'), r'foo/ba .r') + + def test_extract_hackers_arcnames_common_cases(self): + common_hacknames = [ + ('../foo/bar', 'foo/bar'), + ('foo/../bar', 'foo/bar'), + ('foo/../../bar', 'foo/bar'), + ('foo/bar/..', 'foo/bar'), + ('./../foo/bar', 'foo/bar'), + ('/foo/bar', 'foo/bar'), + ('/foo/../bar', 'foo/bar'), + ('/foo/../../bar', 'foo/bar'), + ] + self._test_extract_hackers_arcnames(common_hacknames) + + @unittest.skipIf(os.path.sep != '\\', 'Requires \\ as path separator.') + def test_extract_hackers_arcnames_windows_only(self): + """Test combination of path fixing and windows name sanitization.""" + windows_hacknames = [ + (r'..\foo\bar', 'foo/bar'), + (r'..\/foo\/bar', 'foo/bar'), + (r'foo/\..\/bar', 'foo/bar'), + (r'foo\/../\bar', 'foo/bar'), + (r'C:foo/bar', 'foo/bar'), + (r'C:/foo/bar', 'foo/bar'), + (r'C://foo/bar', 'foo/bar'), + (r'C:\foo\bar', 'foo/bar'), + (r'//conky/mountpoint/foo/bar', 'foo/bar'), + (r'\\conky\mountpoint\foo\bar', 'foo/bar'), + (r'///conky/mountpoint/foo/bar', 'mountpoint/foo/bar'), + (r'\\\conky\mountpoint\foo\bar', 'mountpoint/foo/bar'), + (r'//conky//mountpoint/foo/bar', 'mountpoint/foo/bar'), + (r'\\conky\\mountpoint\foo\bar', 'mountpoint/foo/bar'), + (r'//?/C:/foo/bar', 'foo/bar'), + (r'\\?\C:\foo\bar', 'foo/bar'), + (r'C:/../C:/foo/bar', 'C_/foo/bar'), + (r'a:b\ce|f"g?h*i', 'b/c_d_e_f_g_h_i'), + ('../../foo../../ba..r', 'foo/ba..r'), + ] + self._test_extract_hackers_arcnames(windows_hacknames) + + @unittest.skipIf(os.path.sep != '/', r'Requires / as path separator.') + def test_extract_hackers_arcnames_posix_only(self): + posix_hacknames = [ + ('//foo/bar', 'foo/bar'), + ('../../foo../../ba..r', 'foo../ba..r'), + (r'foo/..\bar', r'foo/..\bar'), + ] + self._test_extract_hackers_arcnames(posix_hacknames) + + def _test_extract_hackers_arcnames(self, hacknames): + for arcname, fixedname in hacknames: + content = b'foobar' + arcname.encode() + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipfp: + zinfo = zipfile.ZipInfo() + # preserve backslashes + zinfo.filename = arcname + zinfo.external_attr = 0o600 << 16 + zipfp.writestr(zinfo, content) + + arcname = arcname.replace(os.sep, "/") + targetpath = os.path.join('target', 'subdir', 'subsub') + correctfile = os.path.join(targetpath, *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname, targetpath) + self.assertEqual(writtenfile, correctfile, + msg='extract %r: %r != %r' % + (arcname, writtenfile, correctfile)) + self.check_file(correctfile, content) + rmtree('target') + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall(targetpath) + self.check_file(correctfile, content) + rmtree('target') + + correctfile = os.path.join(os.getcwd(), *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname) + self.assertEqual(writtenfile, correctfile, + msg="extract %r" % arcname) + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall() + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + unlink(TESTFN2) + + +class OverwriteTests(archiver_tests.OverwriteTests, unittest.TestCase): + testdir = TESTFN + + @classmethod + def setUpClass(cls): + p = cls.ar_with_file = TESTFN + '-with-file.zip' + cls.addClassCleanup(unlink, p) + with zipfile2.ZipFile(p, 'w') as zipfp: + zipfp.writestr('test', b'newcontent') + + p = cls.ar_with_dir = TESTFN + '-with-dir.zip' + cls.addClassCleanup(unlink, p) + with zipfile2.ZipFile(p, 'w') as zipfp: + zipfp.mkdir('test') + + p = cls.ar_with_implicit_dir = TESTFN + '-with-implicit-dir.zip' + cls.addClassCleanup(unlink, p) + with zipfile2.ZipFile(p, 'w') as zipfp: + zipfp.writestr('test/file', b'newcontent') + + def open(self, path): + return zipfile2.ZipFile(path, 'r') + + def extractall(self, ar): + ar.extractall(self.testdir) + + @skip_unless_symlink + def test_overwrite_file_symlink_as_file(self): + self.skipTest('Zipfile2 will overwrite symlinks with a real file') + + @skip_unless_symlink + def test_overwrite_broken_file_symlink_as_file(self): + self.skipTest('Zipfile2 will overwrite symlinks with a real file') + + @unittest.skipIf(sys.platform == 'darwin', 'FIXME: not sure why this test fails') + def test_overwrite_dir_as_file(self): + super().test_overwrite_dir_as_file() + + +class OtherTests(unittest.TestCase): + def test_open_via_zip_info(self): + # Create the ZIP archive + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED, low_level=True) as zipfp: + zipfp.writestr("name", "foo") + with self.assertWarns(UserWarning): + zipfp.writestr("name", "bar") + self.assertEqual(zipfp.namelist(), ["name"] * 2) + + with zipfile2.ZipFile(TESTFN2, "r", low_level=True) as zipfp: + infos = zipfp.infolist() + data = b"" + for info in infos: + with zipfp.open(info) as zipopen: + data += zipopen.read() + self.assertIn(data, {b"foobar", b"barfoo"}) + data = b"" + for info in infos: + data += zipfp.read(info) + self.assertIn(data, {b"foobar", b"barfoo"}) + + def test_writestr_extended_local_header_issue1202(self): + with zipfile2.ZipFile(TESTFN2, 'w') as orig_zip: + for data in 'abcdefghijklmnop': + zinfo = zipfile.ZipInfo(data) + zinfo.flag_bits |= zipfile._MASK_USE_DATA_DESCRIPTOR # Include an extended local header. + orig_zip.writestr(zinfo, data) + + def test_close(self): + """Check that the zipfile is closed after the 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + def test_close_on_exception(self): + """Check that the zipfile is closed if an exception is raised in the + 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + + try: + with zipfile2.ZipFile(TESTFN2, "r") as zipfp2: + raise zipfile.BadZipFile() + except zipfile.BadZipFile: + self.assertIsNone(zipfp2.fp, 'zipfp is not closed') + + def test_unsupported_version(self): + # File has an extract_version of 120 + data = (b'PK\x03\x04x\x00\x00\x00\x00\x00!p\xa1@\x00\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x01\x00\x00\x00xPK\x01\x02x\x03x\x00\x00\x00\x00' + b'\x00!p\xa1@\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x01\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00\x00xPK\x05\x06' + b'\x00\x00\x00\x00\x01\x00\x01\x00/\x00\x00\x00\x1f\x00\x00\x00\x00\x00') + + self.assertRaises(NotImplementedError, zipfile2.ZipFile, + io.BytesIO(data), 'r') + + @requires_zlib() + def test_read_unicode_filenames(self): + # bug #10801 + fname = findfile('zip_cp437_header.zip') + with zipfile2.ZipFile(fname) as zipfp: + for name in zipfp.namelist(): + zipfp.open(name).close() + + def test_write_unicode_filenames(self): + with zipfile2.ZipFile(TESTFN, "w") as zf: + zf.writestr("foo.txt", "Test for unicode filename") + zf.writestr("\xf6.txt", "Test for unicode filename") + self.assertIsInstance(zf.infolist()[0].filename, str) + + with zipfile2.ZipFile(TESTFN, "r") as zf: + self.assertEqual(zf.filelist[0].filename, "foo.txt") + self.assertEqual(zf.filelist[1].filename, "\xf6.txt") + + def create_zipfile_with_extra_data(self, filename, extra_data_name): + with zipfile2.ZipFile(TESTFN, mode='w') as zf: + filename_encoded = filename.encode("utf-8") + # create a ZipInfo object with Unicode path extra field + zip_info = zipfile.ZipInfo(filename) + + tag_for_unicode_path = b'\x75\x70' + version_of_unicode_path = b'\x01' + + import zlib + filename_crc = struct.pack(' os.path.getsize(TESTFN)) + with zipfile2.ZipFile(TESTFN,mode="r") as zipf: + self.assertEqual(zipf.comment, b"shorter comment") + + def test_unicode_comment(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with self.assertRaises(TypeError): + zipf.comment = "this is an error" + + def test_change_comment_in_empty_archive(self): + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertFalse(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_change_comment_in_nonempty_archive(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertTrue(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_empty_zipfile(self): + # Check that creating a file in 'w' or 'a' mode and closing without + # adding any files to the archives creates a valid empty ZIP file + zipf = zipfile2.ZipFile(TESTFN, mode="w") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except zipfile.BadZipFile: + self.fail("Unable to create empty ZIP file in 'w' mode") + + zipf = zipfile2.ZipFile(TESTFN, mode="a") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except: + self.fail("Unable to create empty ZIP file in 'a' mode") + + def test_open_empty_file(self): + # Issue 1710703: Check that opening a file with less than 22 bytes + # raises a BadZipFile exception (rather than the previously unhelpful + # OSError) + f = open(TESTFN, 'w', encoding='utf-8') + f.close() + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN, 'r') + + def test_create_zipinfo_before_1980(self): + self.assertRaises(ValueError, + zipfile.ZipInfo, 'seventies', (1979, 1, 1, 0, 0, 0)) + + def test_create_empty_zipinfo_repr(self): + """Before bpo-26185, repr() on empty ZipInfo object was failing.""" + zi = zipfile.ZipInfo(filename="empty") + self.assertEqual(repr(zi), "") + + def test_create_empty_zipinfo_default_attributes(self): + """Ensure all required attributes are set.""" + zi = zipfile.ZipInfo() + self.assertEqual(zi.orig_filename, "NoName") + self.assertEqual(zi.filename, "NoName") + self.assertEqual(zi.date_time, (1980, 1, 1, 0, 0, 0)) + self.assertEqual(zi.compress_type, zipfile.ZIP_STORED) + self.assertEqual(zi.comment, b"") + self.assertEqual(zi.extra, b"") + self.assertIn(zi.create_system, (0, 3)) + self.assertEqual(zi.create_version, zipfile.DEFAULT_VERSION) + self.assertEqual(zi.extract_version, zipfile.DEFAULT_VERSION) + self.assertEqual(zi.reserved, 0) + self.assertEqual(zi.flag_bits, 0) + self.assertEqual(zi.volume, 0) + self.assertEqual(zi.internal_attr, 0) + self.assertEqual(zi.external_attr, 0) + + # Before bpo-26185, both were missing + self.assertEqual(zi.file_size, 0) + self.assertEqual(zi.compress_size, 0) + + def test_zipfile_with_short_extra_field(self): + """If an extra field in the header is less than 4 bytes, skip it.""" + zipdata = ( + b'PK\x03\x04\x14\x00\x00\x00\x00\x00\x93\x9b\xad@\x8b\x9e' + b'\xd9\xd3\x01\x00\x00\x00\x01\x00\x00\x00\x03\x00\x03\x00ab' + b'c\x00\x00\x00APK\x01\x02\x14\x03\x14\x00\x00\x00\x00' + b'\x00\x93\x9b\xad@\x8b\x9e\xd9\xd3\x01\x00\x00\x00\x01\x00\x00' + b'\x00\x03\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\xa4\x81\x00' + b'\x00\x00\x00abc\x00\x00PK\x05\x06\x00\x00\x00\x00' + b'\x01\x00\x01\x003\x00\x00\x00%\x00\x00\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(zipdata), 'r') as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertIsNone(zipf.testzip()) + + def test_open_conflicting_handles(self): + # It's only possible to open one writable file handle at a time + msg1 = b"It's fun to charter an accountant!" + msg2 = b"And sail the wide accountant sea" + msg3 = b"To find, explore the funds offshore" + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipf: + with zipf.open('foo', mode='w') as w2: + w2.write(msg1) + with zipf.open('bar', mode='w') as w1: + with self.assertRaises(ValueError): + zipf.open('handle', mode='w') + with self.assertRaises(ValueError): + zipf.open('foo', mode='r') + with self.assertRaises(ValueError): + zipf.writestr('str', 'abcde') + with self.assertRaises(ValueError): + zipf.write(__file__, 'file') + with self.assertRaises(ValueError): + zipf.close() + w1.write(msg2) + with zipf.open('baz', mode='w') as w2: + w2.write(msg3) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipf: + self.assertEqual(zipf.read('foo'), msg1) + self.assertEqual(zipf.read('bar'), msg2) + self.assertEqual(zipf.read('baz'), msg3) + self.assertEqual(zipf.namelist(), ['foo', 'bar', 'baz']) + + def test_seek_tell(self): + # Test seek functionality + txt = b"Where's Bruce?" + bloc = txt.find(b"Bruce") + # Check seek on a file + with zipfile2.ZipFile(TESTFN, "w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(TESTFN, "r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + self.assertEqual(fp.tell(), bloc + 5) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + # Check seek on memory file + data = io.BytesIO() + with zipfile2.ZipFile(data, mode="w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(data, mode="r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + self.assertEqual(fp.tell(), bloc + 5) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + + def test_read_after_seek(self): + # Issue 102956: Make sure seek(x, os.SEEK_CUR) doesn't break read() + txt = b"Charge men!" + bloc = txt.find(b"men") + with zipfile2.ZipFile(TESTFN, "w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(TESTFN, mode="r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.read(-1), b'men!') + with zipfile2.ZipFile(TESTFN, mode="r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.read(6) + fp.seek(1, os.SEEK_CUR) + self.assertEqual(fp.read(-1), b'men!') + + @requires_bz2() + def test_decompress_without_3rd_party_library(self): + data = b'PK\x05\x06\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + zip_file = io.BytesIO(data) + with zipfile2.ZipFile(zip_file, 'w', compression=zipfile.ZIP_BZIP2) as zf: + zf.writestr('a.txt', b'a') + with mock.patch('zipfile.bz2', None): + with zipfile2.ZipFile(zip_file) as zf: + self.assertRaises(RuntimeError, zf.extract, 'a.txt') + + @requires_zlib() + def test_full_overlap(self): + data = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e' + b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00a\xed' + b'\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\d\x0b`P' + b'K\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2' + b'\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00aPK' + b'\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e' + b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00bPK\x05' + b'\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00\x00/\x00\x00' + b'\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertEqual(zipf.namelist(), ['a', 'b']) + zi = zipf.getinfo('a') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + zi = zipf.getinfo('b') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + self.assertEqual(len(zipf.read('a')), 1033) + with self.assertRaisesRegex(zipfile.BadZipFile, 'File name.*differ'): + zipf.read('b') + + @requires_zlib() + def test_quoted_overlap(self): + data = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05Y\xfc' + b'8\x044\x00\x00\x00(\x04\x00\x00\x01\x00\x00\x00a\x00' + b'\x1f\x00\xe0\xffPK\x03\x04\x14\x00\x00\x00\x08\x00\xa0l' + b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00' + b'\x00\x00b\xed\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\' + b'd\x0b`PK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0' + b'lH\x05Y\xfc8\x044\x00\x00\x00(\x04\x00\x00\x01' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x00aPK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0l' + b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00$\x00\x00\x00' + b'bPK\x05\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00' + b'\x00S\x00\x00\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertEqual(zipf.namelist(), ['a', 'b']) + zi = zipf.getinfo('a') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 52) + self.assertEqual(zi.file_size, 1064) + zi = zipf.getinfo('b') + self.assertEqual(zi.header_offset, 36) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + with self.assertRaisesRegex(zipfile.BadZipFile, 'Overlapped entries'): + zipf.read('a') + self.assertEqual(len(zipf.read('b')), 1033) + + def tearDown(self): + unlink(TESTFN) + unlink(TESTFN2) + + +class AbstractBadCrcTests: + def test_testzip_with_bad_crc(self): + """Tests that files with bad CRCs return their name from testzip.""" + zipdata = self.zip_with_bad_crc + + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertEqual('afile', zipf.testzip()) + + def test_read_with_bad_crc(self): + """Tests that files with bad CRCs raise a BadZipFile exception when read.""" + zipdata = self.zip_with_bad_crc + + # Using ZipFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + self.assertRaises(zipfile.BadZipFile, zipf.read, 'afile') + + # Using ZipExtFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + self.assertRaises(zipfile.BadZipFile, corrupt_file.read) + + # Same with small reads (in order to exercise the buffering logic) + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + corrupt_file.MIN_READ_SIZE = 2 + with self.assertRaises(zipfile.BadZipFile): + while corrupt_file.read(2): + pass + + +class StoredBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_STORED + zip_with_bad_crc = ( + b'PK\003\004\024\0\0\0\0\0 \213\212;:r' + b'\253\377\f\0\0\0\f\0\0\0\005\0\0\000af' + b'ilehello,AworldP' + b'K\001\002\024\003\024\0\0\0\0\0 \213\212;:' + b'r\253\377\f\0\0\0\f\0\0\0\005\0\0\0\0' + b'\0\0\0\0\0\0\0\200\001\0\0\0\000afi' + b'lePK\005\006\0\0\0\0\001\0\001\0003\000' + b'\0\0/\0\0\0\0\0') + +@requires_zlib() +class DeflateBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_DEFLATED + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00n}\x0c=FA' + b'KE\x10\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\xcbH\xcd\xc9\xc9W(\xcf/\xcaI\xc9\xa0' + b'=\x13\x00PK\x01\x02\x14\x03\x14\x00\x00\x00\x08\x00n' + b'}\x0c=FAKE\x10\x00\x00\x00n\x00\x00\x00\x05' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00' + b'\x00afilePK\x05\x06\x00\x00\x00\x00\x01\x00' + b'\x01\x003\x00\x00\x003\x00\x00\x00\x00\x00') + +@requires_bz2() +class Bzip2BadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_BZIP2 + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0c\x00nu\x0c=FA' + b'KE8\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ileBZh91AY&SY\xd4\xa8\xca' + b'\x7f\x00\x00\x0f\x11\x80@\x00\x06D\x90\x80 \x00 \xa5' + b'P\xd9!\x03\x03\x13\x13\x13\x89\xa9\xa9\xc2u5:\x9f' + b'\x8b\xb9"\x9c(HjTe?\x80PK\x01\x02\x14' + b'\x03\x14\x03\x00\x00\x0c\x00nu\x0c=FAKE8' + b'\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x00 \x80\x80\x81\x00\x00\x00\x00afilePK' + b'\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00\x00[\x00' + b'\x00\x00\x00\x00') + +@requires_lzma() +class LzmaBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_LZMA + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\t\x04\x05\x00]\x00\x00\x00\x04\x004\x19I' + b'\xee\x8d\xe9\x17\x89:3`\tq!.8\x00PK' + b'\x01\x02\x14\x03\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00 \x80\x80\x81\x00\x00\x00\x00afil' + b'ePK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00' + b'\x00>\x00\x00\x00\x00\x00') + + +class DecryptionTests(unittest.TestCase): + """Check that ZIP decryption works. Since the library does not + support encryption at the moment, we use a pre-generated encrypted + ZIP file.""" + + data = ( + b'PK\x03\x04\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00\x1a\x00' + b'\x00\x00\x08\x00\x00\x00test.txt\xfa\x10\xa0gly|\xfa-\xc5\xc0=\xf9y' + b'\x18\xe0\xa8r\xb3Z}Lg\xbc\xae\xf9|\x9b\x19\xe4\x8b\xba\xbb)\x8c\xb0\xdbl' + b'PK\x01\x02\x14\x00\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00' + b'\x1a\x00\x00\x00\x08\x00\x00\x00\x00\x00\x00\x00\x01\x00 \x00\xb6\x81' + b'\x00\x00\x00\x00test.txtPK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x006\x00' + b'\x00\x00L\x00\x00\x00\x00\x00' ) + data2 = ( + b'PK\x03\x04\x14\x00\t\x00\x08\x00\xcf}38xu\xaa\xb2\x14\x00\x00\x00\x00\x02' + b'\x00\x00\x04\x00\x15\x00zeroUT\t\x00\x03\xd6\x8b\x92G\xda\x8b\x92GUx\x04' + b'\x00\xe8\x03\xe8\x03\xc7", "exec") +raise_src = 'def do_raise(): raise TypeError\n' + +def make_pyc(co, mtime, size): + data = marshal.dumps(co) + pyc = (importlib.util.MAGIC_NUMBER + + struct.pack("", "exec"), NOW, len(src)) + files = {TESTMOD + pyc_ext: pyc, + "some.data": "some data"} + self.doTest(pyc_ext, files, TESTMOD, prefix='') + + def testDefaultOptimizationLevel(self): + # zipimport should use the default optimization level (#28131) + src = """if 1: # indent hack + def test(val): + assert(val) + return val\n""" + files = {TESTMOD + '.py': src} + self.makeZip(files) + sys.path.insert(0, TEMP_ZIP) + mod = importlib.import_module(TESTMOD) + self.assertEqual(mod.test(1), 1) + if __debug__: + self.assertRaises(AssertionError, mod.test, False) + else: + self.assertEqual(mod.test(0), 0) + + def testImport_WithStuff(self): + # try importing from a zipfile which contains additional + # stuff at the beginning of the file + files = {TESTMOD + ".py": test_src} + self.doTest(".py", files, TESTMOD, + stuff=b"Some Stuff"*31) + + def assertModuleSource(self, module): + self.assertEqual(inspect.getsource(module), test_src) + + def testGetSource(self): + files = {TESTMOD + ".py": test_src} + self.doTest(".py", files, TESTMOD, call=self.assertModuleSource) + + def testGetCompiledSource(self): + pyc = make_pyc(compile(test_src, "", "exec"), NOW, len(test_src)) + files = {TESTMOD + ".py": test_src, + TESTMOD + pyc_ext: pyc} + self.doTest(pyc_ext, files, TESTMOD, call=self.assertModuleSource) + + def runDoctest(self, callback): + files = {TESTMOD + ".py": test_src, + "xyz.txt": ">>> log.append(True)\n"} + self.doTest(".py", files, TESTMOD, call=callback) + + def doDoctestFile(self, module): + log = [] + old_master, doctest.master = doctest.master, None + try: + doctest.testfile( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ) + finally: + doctest.master = old_master + self.assertEqual(log,[True]) + + def testDoctestFile(self): + self.runDoctest(self.doDoctestFile) + + def doDoctestSuite(self, module): + log = [] + doctest.DocFileTest( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ).run() + self.assertEqual(log,[True]) + + def testDoctestSuite(self): + self.runDoctest(self.doDoctestSuite) + + def doTraceback(self, module): + try: + module.do_raise() + except Exception as e: + tb = e.__traceback__.tb_next + + f,lno,n,line = extract_tb(tb, 1)[0] + self.assertEqual(line, raise_src.strip()) + + f,lno,n,line = extract_stack(tb.tb_frame, 1)[0] + self.assertEqual(line, raise_src.strip()) + + s = io.StringIO() + print_tb(tb, 1, s) + self.assertTrue(s.getvalue().endswith( + ' def do_raise(): raise TypeError\n' + '' if support.has_no_debug_ranges() else + ' ^^^^^^^^^^^^^^^\n' + )) + else: + raise AssertionError("This ought to be impossible") + + def testTraceback(self): + files = {TESTMOD + ".py": raise_src} + self.doTest(None, files, TESTMOD, call=self.doTraceback) + + @unittest.skipIf(os_helper.TESTFN_UNENCODABLE is None, + "need an unencodable filename") + def testUnencodable(self): + filename = os_helper.TESTFN_UNENCODABLE + ".zip" + self.makeZip({TESTMOD + ".py": test_src}, filename) + spec = zipimport.zipimporter(filename).find_spec(TESTMOD) + mod = importlib.util.module_from_spec(spec) + spec.loader.exec_module(mod) + + def testBytesPath(self): + filename = os_helper.TESTFN + ".zip" + self.makeZip({TESTMOD + ".py": test_src}, filename) + + zipimport.zipimporter(filename) + with self.assertRaises(TypeError): + zipimport.zipimporter(os.fsencode(filename)) + with self.assertRaises(TypeError): + zipimport.zipimporter(bytearray(os.fsencode(filename))) + with self.assertRaises(TypeError): + zipimport.zipimporter(memoryview(os.fsencode(filename))) + + def testComment(self): + files = {TESTMOD + ".py": test_src} + self.doTest(".py", files, TESTMOD, comment=b"comment") + + def testBeginningCruftAndComment(self): + files = {TESTMOD + ".py": test_src} + self.doTest(".py", files, TESTMOD, stuff=b"cruft" * 64, comment=b"hi") + + def testLargestPossibleComment(self): + files = {TESTMOD + ".py": test_src} + self.doTest(".py", files, TESTMOD, comment=b"c" * ((1 << 16) - 1)) + + +@support.requires_zlib() +class CompressedZipImportTestCase(UncompressedZipImportTestCase): + compression = ZIP_DEFLATED + + +class BadFileZipImportTestCase(unittest.TestCase): + def assertZipFailure(self, filename): + self.assertRaises(zipimport.ZipImportError, + zipimport.zipimporter, filename) + + def testNoFile(self): + self.assertZipFailure('AdfjdkFJKDFJjdklfjs') + + def testEmptyFilename(self): + self.assertZipFailure('') + + def testBadArgs(self): + self.assertRaises(TypeError, zipimport.zipimporter, None) + self.assertRaises(TypeError, zipimport.zipimporter, TESTMOD, kwd=None) + self.assertRaises(TypeError, zipimport.zipimporter, + list(os.fsencode(TESTMOD))) + + def testFilenameTooLong(self): + self.assertZipFailure('A' * 33000) + + def testEmptyFile(self): + os_helper.unlink(TESTMOD) + os_helper.create_empty_file(TESTMOD) + self.assertZipFailure(TESTMOD) + + @unittest.skipIf(support.is_wasi, "mode 000 not supported.") + def testFileUnreadable(self): + os_helper.unlink(TESTMOD) + fd = os.open(TESTMOD, os.O_CREAT, 000) + try: + os.close(fd) + + with self.assertRaises(zipimport.ZipImportError) as cm: + zipimport.zipimporter(TESTMOD) + finally: + # If we leave "the read-only bit" set on Windows, nothing can + # delete TESTMOD, and later tests suffer bogus failures. + os.chmod(TESTMOD, 0o666) + os_helper.unlink(TESTMOD) + + def testNotZipFile(self): + os_helper.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write('a' * 22) + fp.close() + self.assertZipFailure(TESTMOD) + + # XXX: disabled until this works on Big-endian machines + def _testBogusZipFile(self): + os_helper.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write(struct.pack('=I', 0x06054B50)) + fp.write('a' * 18) + fp.close() + z = zipimport.zipimporter(TESTMOD) + + try: + with warnings.catch_warnings(): + warnings.simplefilter("ignore", DeprecationWarning) + self.assertRaises(TypeError, z.load_module, None) + self.assertRaises(TypeError, z.find_module, None) + self.assertRaises(TypeError, z.find_spec, None) + self.assertRaises(TypeError, z.exec_module, None) + self.assertRaises(TypeError, z.is_package, None) + self.assertRaises(TypeError, z.get_code, None) + self.assertRaises(TypeError, z.get_data, None) + self.assertRaises(TypeError, z.get_source, None) + + error = zipimport.ZipImportError + self.assertIsNone(z.find_spec('abc')) + + with warnings.catch_warnings(): + warnings.simplefilter("ignore", DeprecationWarning) + self.assertRaises(error, z.load_module, 'abc') + self.assertRaises(error, z.get_code, 'abc') + self.assertRaises(OSError, z.get_data, 'abc') + self.assertRaises(error, z.get_source, 'abc') + self.assertRaises(error, z.is_package, 'abc') + finally: + zipimport._zip_directory_cache.clear() + + +def tearDownModule(): + os_helper.unlink(TESTMOD) + + +if __name__ == "__main__": + unittest.main() diff --git a/cpython-tests/3.12/zip_cp437_header.zip b/cpython-tests/3.12/zip_cp437_header.zip new file mode 100644 index 0000000..f7c6cf1 Binary files /dev/null and b/cpython-tests/3.12/zip_cp437_header.zip differ diff --git a/cpython-tests/3.12/zipdir.zip b/cpython-tests/3.12/zipdir.zip new file mode 100644 index 0000000..ac21d7a Binary files /dev/null and b/cpython-tests/3.12/zipdir.zip differ diff --git a/cpython-tests/3.12/zipdir_backslash.zip b/cpython-tests/3.12/zipdir_backslash.zip new file mode 100644 index 0000000..979126e Binary files /dev/null and b/cpython-tests/3.12/zipdir_backslash.zip differ diff --git a/cpython-tests/3.12/ziptestdata/README.md b/cpython-tests/3.12/ziptestdata/README.md new file mode 100644 index 0000000..00d96d4 --- /dev/null +++ b/cpython-tests/3.12/ziptestdata/README.md @@ -0,0 +1,35 @@ +# Test data for `test_zipfile` + +The test executables in this directory are created manually from header.sh and +the `testdata_module_inside_zip.py` file. You must have Info-ZIP's zip utility +installed (`apt install zip` on Debian). + +## Purpose + +These are used to test executable files with an appended zipfile, in a scenario +where the executable is _not_ a Python interpreter itself so our automatic +zipimport machinery (that'd look for `__main__.py`) is not being used. + +## Updating the test executables + +If you update header.sh or the testdata_module_inside_zip.py file, rerun the +commands below. These are expected to be rarely changed, if ever. + +### Standard old format (2.0) zip file + +``` +zip -0 zip2.zip testdata_module_inside_zip.py +cat header.sh zip2.zip >exe_with_zip +rm zip2.zip +``` + +### Modern format (4.5) zip64 file + +Redirecting from stdin forces Info-ZIP's zip tool to create a zip64. + +``` +zip -0 zip64.zip +cat header.sh zip64.zip >exe_with_z64 +rm zip64.zip +``` + diff --git a/cpython-tests/3.12/ziptestdata/exe_with_z64 b/cpython-tests/3.12/ziptestdata/exe_with_z64 new file mode 100644 index 0000000..82b03cf Binary files /dev/null and b/cpython-tests/3.12/ziptestdata/exe_with_z64 differ diff --git a/cpython-tests/3.12/ziptestdata/exe_with_zip b/cpython-tests/3.12/ziptestdata/exe_with_zip new file mode 100644 index 0000000..c833cdf Binary files /dev/null and b/cpython-tests/3.12/ziptestdata/exe_with_zip differ diff --git a/cpython-tests/3.12/ziptestdata/header.sh b/cpython-tests/3.12/ziptestdata/header.sh new file mode 100644 index 0000000..52dc91a --- /dev/null +++ b/cpython-tests/3.12/ziptestdata/header.sh @@ -0,0 +1,24 @@ +#!/bin/bash +INTERPRETER_UNDER_TEST="$1" +if [[ ! -x "${INTERPRETER_UNDER_TEST}" ]]; then + echo "Interpreter must be the command line argument." + exit 4 +fi +EXECUTABLE="$0" exec "${INTERPRETER_UNDER_TEST}" -E - <e|f"g?h*i', ','), r'a\b,c_d_e_f_g_h_i') + self.assertEqual(san('../../foo../../ba..r', '/'), r'foo/ba..r') + + def test_extract_hackers_arcnames_common_cases(self): + common_hacknames = [ + ('../foo/bar', 'foo/bar'), + ('foo/../bar', 'foo/bar'), + ('foo/../../bar', 'foo/bar'), + ('foo/bar/..', 'foo/bar'), + ('./../foo/bar', 'foo/bar'), + ('/foo/bar', 'foo/bar'), + ('/foo/../bar', 'foo/bar'), + ('/foo/../../bar', 'foo/bar'), + ] + self._test_extract_hackers_arcnames(common_hacknames) + + @unittest.skipIf(os.path.sep != '\\', 'Requires \\ as path separator.') + def test_extract_hackers_arcnames_windows_only(self): + """Test combination of path fixing and windows name sanitization.""" + windows_hacknames = [ + (r'..\foo\bar', 'foo/bar'), + (r'..\/foo\/bar', 'foo/bar'), + (r'foo/\..\/bar', 'foo/bar'), + (r'foo\/../\bar', 'foo/bar'), + (r'C:foo/bar', 'foo/bar'), + (r'C:/foo/bar', 'foo/bar'), + (r'C://foo/bar', 'foo/bar'), + (r'C:\foo\bar', 'foo/bar'), + (r'//conky/mountpoint/foo/bar', 'foo/bar'), + (r'\\conky\mountpoint\foo\bar', 'foo/bar'), + (r'///conky/mountpoint/foo/bar', 'conky/mountpoint/foo/bar'), + (r'\\\conky\mountpoint\foo\bar', 'conky/mountpoint/foo/bar'), + (r'//conky//mountpoint/foo/bar', 'conky/mountpoint/foo/bar'), + (r'\\conky\\mountpoint\foo\bar', 'conky/mountpoint/foo/bar'), + (r'//?/C:/foo/bar', 'foo/bar'), + (r'\\?\C:\foo\bar', 'foo/bar'), + (r'C:/../C:/foo/bar', 'C_/foo/bar'), + (r'a:b\ce|f"g?h*i', 'b/c_d_e_f_g_h_i'), + ('../../foo../../ba..r', 'foo/ba..r'), + ] + self._test_extract_hackers_arcnames(windows_hacknames) + + @unittest.skipIf(os.path.sep != '/', r'Requires / as path separator.') + def test_extract_hackers_arcnames_posix_only(self): + posix_hacknames = [ + ('//foo/bar', 'foo/bar'), + ('../../foo../../ba..r', 'foo../ba..r'), + (r'foo/..\bar', r'foo/..\bar'), + ] + self._test_extract_hackers_arcnames(posix_hacknames) + + def _test_extract_hackers_arcnames(self, hacknames): + for arcname, fixedname in hacknames: + content = b'foobar' + arcname.encode() + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipfp: + zinfo = zipfile.ZipInfo() + # preserve backslashes + zinfo.filename = arcname + zinfo.external_attr = 0o600 << 16 + zipfp.writestr(zinfo, content) + + arcname = arcname.replace(os.sep, "/") + targetpath = os.path.join('target', 'subdir', 'subsub') + correctfile = os.path.join(targetpath, *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname, targetpath) + self.assertEqual(writtenfile, correctfile, + msg='extract %r: %r != %r' % + (arcname, writtenfile, correctfile)) + self.check_file(correctfile, content) + rmtree('target') + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall(targetpath) + self.check_file(correctfile, content) + rmtree('target') + + correctfile = os.path.join(os.getcwd(), *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname) + self.assertEqual(writtenfile, correctfile, + msg="extract %r" % arcname) + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall() + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + unlink(TESTFN2) + + +class OtherTests(unittest.TestCase): + + def test_open_via_zip_info(self): + # Create the ZIP archive + with zipfile2.ZipFile( + TESTFN2, "w", zipfile.ZIP_STORED, low_level=True) as zipfp: + zipfp.writestr("name", "foo") + with self.assertWarns(UserWarning): + zipfp.writestr("name", "bar") + self.assertEqual(zipfp.namelist(), ["name"] * 2) + + with zipfile2.ZipFile(TESTFN2, "r", low_level=True) as zipfp: + infos = zipfp.infolist() + data = b"" + for info in infos: + with zipfp.open(info) as zipopen: + data += zipopen.read() + self.assertIn(data, {b"foobar", b"barfoo"}) + data = b"" + for info in infos: + data += zipfp.read(info) + self.assertIn(data, {b"foobar", b"barfoo"}) + + def test_writestr_extended_local_header_issue1202(self): + with zipfile2.ZipFile(TESTFN2, 'w') as orig_zip: + for data in 'abcdefghijklmnop': + zinfo = zipfile.ZipInfo(data) + zinfo.flag_bits |= 0x08 # Include an extended local header. + orig_zip.writestr(zinfo, data) + + def test_close(self): + """Check that the zipfile is closed after the 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + def test_close_on_exception(self): + """Check that the zipfile is closed if an exception is raised in the + 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + + try: + with zipfile2.ZipFile(TESTFN2, "r") as zipfp2: + raise zipfile.BadZipFile() + except zipfile.BadZipFile: + self.assertIsNone(zipfp2.fp, 'zipfp is not closed') + + def test_unsupported_version(self): + # File has an extract_version of 120 + data = (b'PK\x03\x04x\x00\x00\x00\x00\x00!p\xa1@\x00\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x01\x00\x00\x00xPK\x01\x02x\x03x\x00\x00\x00\x00' + b'\x00!p\xa1@\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x01\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00\x00xPK\x05\x06' + b'\x00\x00\x00\x00\x01\x00\x01\x00/\x00\x00\x00\x1f\x00\x00\x00\x00\x00') + + self.assertRaises(NotImplementedError, zipfile2.ZipFile, + io.BytesIO(data), 'r') + + @requires_zlib + def test_read_unicode_filenames(self): + # bug #10801 + fname = findfile('zip_cp437_header.zip') + with zipfile2.ZipFile(fname) as zipfp: + for name in zipfp.namelist(): + zipfp.open(name).close() + + def test_write_unicode_filenames(self): + with zipfile2.ZipFile(TESTFN, "w") as zf: + zf.writestr("foo.txt", "Test for unicode filename") + zf.writestr("\xf6.txt", "Test for unicode filename") + self.assertIsInstance(zf.infolist()[0].filename, str) + + with zipfile2.ZipFile(TESTFN, "r") as zf: + self.assertEqual(zf.filelist[0].filename, "foo.txt") + self.assertEqual(zf.filelist[1].filename, "\xf6.txt") + + def test_read_after_write_unicode_filenames(self): + with zipfile2.ZipFile(TESTFN2, 'w') as zipfp: + zipfp.writestr('приклад', b'sample') + self.assertEqual(zipfp.read('приклад'), b'sample') + + def test_exclusive_create_zip_file(self): + """Test exclusive creating a new zipfile.""" + unlink(TESTFN2) + filename = 'testfile.txt' + content = b'hello, world. this is some content.' + with zipfile2.ZipFile(TESTFN2, "x", zipfile.ZIP_STORED) as zipfp: + zipfp.writestr(filename, content) + with self.assertRaises(FileExistsError): + zipfile2.ZipFile(TESTFN2, "x", zipfile.ZIP_STORED) + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertEqual(zipfp.namelist(), [filename]) + self.assertEqual(zipfp.read(filename), content) + + def test_create_non_existent_file_for_append(self): + if os.path.exists(TESTFN): + os.unlink(TESTFN) + + filename = 'testfile.txt' + content = b'hello, world. this is some content.' + + try: + with zipfile2.ZipFile(TESTFN, 'a') as zf: + zf.writestr(filename, content) + except OSError: + self.fail('Could not append data to a non-existent zip file.') + + self.assertTrue(os.path.exists(TESTFN)) + + with zipfile2.ZipFile(TESTFN, 'r') as zf: + self.assertEqual(zf.read(filename), content) + + def test_close_erroneous_file(self): + # This test checks that the ZipFile constructor closes the file object + # it opens if there's an error in the file. If it doesn't, the + # traceback holds a reference to the ZipFile object and, indirectly, + # the file object. + # On Windows, this causes the os.unlink() call to fail because the + # underlying file is still open. This is SF bug #412214. + # + with open(TESTFN, "w") as fp: + fp.write("this is not a legal zip file\n") + try: + zf = zipfile2.ZipFile(TESTFN) + except zipfile.BadZipFile: + pass + + def test_is_zip_erroneous_file(self): + """Check that is_zipfile() correctly identifies non-zip files.""" + # - passing a filename + with open(TESTFN, "w") as fp: + fp.write("this is not a legal zip file\n") + self.assertFalse(zipfile.is_zipfile(TESTFN)) + # - passing a path-like object + self.assertFalse(zipfile.is_zipfile(pathlib.Path(TESTFN))) + # - passing a file object + with open(TESTFN, "rb") as fp: + self.assertFalse(zipfile.is_zipfile(fp)) + # - passing a file-like object + fp = io.BytesIO() + fp.write(b"this is not a legal zip file\n") + self.assertFalse(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + self.assertFalse(zipfile.is_zipfile(fp)) + + def test_damaged_zipfile(self): + """Check that zipfiles with missing bytes at the end raise BadZipFile.""" + # - Create a valid zip file + fp = io.BytesIO() + with zipfile2.ZipFile(fp, mode="w") as zipf: + zipf.writestr("foo.txt", b"O, for a Muse of Fire!") + zipfiledata = fp.getvalue() + + # - Now create copies of it missing the last N bytes and make sure + # a BadZipFile exception is raised when we try to open it + for N in range(len(zipfiledata)): + fp = io.BytesIO(zipfiledata[:N]) + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, fp) + + def test_is_zip_valid_file(self): + """Check that is_zipfile() correctly identifies zip files.""" + # - passing a filename + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt", b"O, for a Muse of Fire!") + + self.assertTrue(zipfile.is_zipfile(TESTFN)) + # - passing a file object + with open(TESTFN, "rb") as fp: + self.assertTrue(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + zip_contents = fp.read() + # - passing a file-like object + fp = io.BytesIO() + fp.write(zip_contents) + self.assertTrue(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + self.assertTrue(zipfile.is_zipfile(fp)) + + def test_non_existent_file_raises_OSError(self): + # make sure we don't raise an AttributeError when a partially-constructed + # ZipFile instance is finalized; this tests for regression on SF tracker + # bug #403871. + + # The bug we're testing for caused an AttributeError to be raised + # when a ZipFile instance was created for a file that did not + # exist; the .fp member was not initialized but was needed by the + # __del__() method. Since the AttributeError is in the __del__(), + # it is ignored, but the user should be sufficiently annoyed by + # the message on the output that regression will be noticed + # quickly. + self.assertRaises(OSError, zipfile2.ZipFile, TESTFN) + + def test_empty_file_raises_BadZipFile(self): + f = open(TESTFN, 'w') + f.close() + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN) + + with open(TESTFN, 'w') as fp: + fp.write("short file") + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN) + + def test_closed_zip_raises_ValueError(self): + """Verify that testzip() doesn't swallow inappropriate exceptions.""" + data = io.BytesIO() + with zipfile2.ZipFile(data, mode="w") as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + + # This is correct; calling .read on a closed ZipFile should raise + # a ValueError, and so should calling .testzip. An earlier + # version of .testzip would swallow this exception (and any other) + # and report that the first file in the archive was corrupt. + self.assertRaises(ValueError, zipf.read, "foo.txt") + self.assertRaises(ValueError, zipf.open, "foo.txt") + self.assertRaises(ValueError, zipf.testzip) + self.assertRaises(ValueError, zipf.writestr, "bogus.txt", "bogus") + with open(TESTFN, 'w') as f: + f.write('zipfile test data') + self.assertRaises(ValueError, zipf.write, TESTFN) + + def test_bad_constructor_mode(self): + """Check that bad modes passed to ZipFile constructor are caught.""" + self.assertRaises(ValueError, zipfile2.ZipFile, TESTFN, "q") + + def test_bad_open_mode(self): + """Check that bad modes passed to ZipFile.open are caught.""" + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + + with zipfile2.ZipFile(TESTFN, mode="r") as zipf: + # read the data to make sure the file is there + zipf.read("foo.txt") + self.assertRaises(ValueError, zipf.open, "foo.txt", "q") + # universal newlines support is removed + self.assertRaises(ValueError, zipf.open, "foo.txt", "U") + self.assertRaises(ValueError, zipf.open, "foo.txt", "rU") + + def test_read0(self): + """Check that calling read(0) on a ZipExtFile object returns an empty + string and doesn't advance file pointer.""" + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + # read the data to make sure the file is there + with zipf.open("foo.txt") as f: + for i in range(FIXEDTEST_SIZE): + self.assertEqual(f.read(0), b'') + + self.assertEqual(f.read(), b"O, for a Muse of Fire!") + + def test_open_non_existent_item(self): + """Check that attempting to call open() for an item that doesn't + exist in the archive raises a RuntimeError.""" + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + self.assertRaises(KeyError, zipf.open, "foo.txt", "r") + + def test_bad_compression_mode(self): + """Check that bad compression methods passed to ZipFile.open are + caught.""" + self.assertRaises(NotImplementedError, zipfile2.ZipFile, TESTFN, "w", -1) + + def test_unsupported_compression(self): + # data is declared as shrunk, but actually deflated + data = (b'PK\x03\x04.\x00\x00\x00\x01\x00\xe4C\xa1@\x00\x00\x00' + b'\x00\x02\x00\x00\x00\x00\x00\x00\x00\x01\x00\x00\x00x\x03\x00PK\x01' + b'\x02.\x03.\x00\x00\x00\x01\x00\xe4C\xa1@\x00\x00\x00\x00\x02\x00\x00' + b'\x00\x00\x00\x00\x00\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x80\x01\x00\x00\x00\x00xPK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x00' + b'/\x00\x00\x00!\x00\x00\x00\x00\x00') + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertRaises(NotImplementedError, zipf.open, 'x') + + def test_null_byte_in_filename(self): + """Check that a filename containing a null byte is properly + terminated.""" + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt\x00qqq", b"O, for a Muse of Fire!") + self.assertEqual(zipf.namelist(), ['foo.txt']) + + def test_struct_sizes(self): + """Check that ZIP internal structure sizes are calculated correctly.""" + self.assertEqual(zipfile.sizeEndCentDir, 22) + self.assertEqual(zipfile.sizeCentralDir, 46) + self.assertEqual(zipfile.sizeEndCentDir64, 56) + self.assertEqual(zipfile.sizeEndCentDir64Locator, 20) + + def test_comments(self): + """Check that comments on the archive are handled properly.""" + + # check default comment is empty + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + self.assertEqual(zipf.comment, b'') + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + + with zipfile2.ZipFile(TESTFN, mode="r") as zipfr: + self.assertEqual(zipfr.comment, b'') + + # check a simple short comment + comment = b'Bravely taking to his feet, he beat a very brave retreat.' + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.comment = comment + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN, mode="r") as zipfr: + self.assertEqual(zipf.comment, comment) + + # check a comment of max length + comment2 = ''.join(['%d' % (i**3 % 10) for i in range((1 << 16)-1)]) + comment2 = comment2.encode("ascii") + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.comment = comment2 + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + + with zipfile2.ZipFile(TESTFN, mode="r") as zipfr: + self.assertEqual(zipfr.comment, comment2) + + # check a comment that is too long is truncated + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + with self.assertWarns(UserWarning): + zipf.comment = comment2 + b'oops' + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN, mode="r") as zipfr: + self.assertEqual(zipfr.comment, comment2) + + # check that comments are correctly modified in append mode + with zipfile2.ZipFile(TESTFN,mode="w") as zipf: + zipf.comment = b"original comment" + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN,mode="a") as zipf: + zipf.comment = b"an updated comment" + with zipfile2.ZipFile(TESTFN,mode="r") as zipf: + self.assertEqual(zipf.comment, b"an updated comment") + + # check that comments are correctly shortened in append mode + # and the file is indeed truncated + with zipfile2.ZipFile(TESTFN,mode="w") as zipf: + zipf.comment = b"original comment that's longer" + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + original_zip_size = os.path.getsize(TESTFN) + with zipfile2.ZipFile(TESTFN,mode="a") as zipf: + zipf.comment = b"shorter comment" + self.assertTrue(original_zip_size > os.path.getsize(TESTFN)) + with zipfile2.ZipFile(TESTFN,mode="r") as zipf: + self.assertEqual(zipf.comment, b"shorter comment") + + def test_unicode_comment(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with self.assertRaises(TypeError): + zipf.comment = "this is an error" + + def test_change_comment_in_empty_archive(self): + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertFalse(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_change_comment_in_nonempty_archive(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertTrue(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_empty_zipfile(self): + # Check that creating a file in 'w' or 'a' mode and closing without + # adding any files to the archives creates a valid empty ZIP file + zipf = zipfile2.ZipFile(TESTFN, mode="w") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except zipfile.BadZipFile: + self.fail("Unable to create empty ZIP file in 'w' mode") + + zipf = zipfile2.ZipFile(TESTFN, mode="a") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except: + self.fail("Unable to create empty ZIP file in 'a' mode") + + def test_open_empty_file(self): + # Issue 1710703: Check that opening a file with less than 22 bytes + # raises a BadZipFile exception (rather than the previously unhelpful + # OSError) + f = open(TESTFN, 'w') + f.close() + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN, 'r') + + def test_create_zipinfo_before_1980(self): + self.assertRaises(ValueError, + zipfile.ZipInfo, 'seventies', (1979, 1, 1, 0, 0, 0)) + + def test_zipfile_with_short_extra_field(self): + """If an extra field in the header is less than 4 bytes, skip it.""" + zipdata = ( + b'PK\x03\x04\x14\x00\x00\x00\x00\x00\x93\x9b\xad@\x8b\x9e' + b'\xd9\xd3\x01\x00\x00\x00\x01\x00\x00\x00\x03\x00\x03\x00ab' + b'c\x00\x00\x00APK\x01\x02\x14\x03\x14\x00\x00\x00\x00' + b'\x00\x93\x9b\xad@\x8b\x9e\xd9\xd3\x01\x00\x00\x00\x01\x00\x00' + b'\x00\x03\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\xa4\x81\x00' + b'\x00\x00\x00abc\x00\x00PK\x05\x06\x00\x00\x00\x00' + b'\x01\x00\x01\x003\x00\x00\x00%\x00\x00\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(zipdata), 'r') as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertIsNone(zipf.testzip()) + + def test_open_conflicting_handles(self): + # It's only possible to open one writable file handle at a time + msg1 = b"It's fun to charter an accountant!" + msg2 = b"And sail the wide accountant sea" + msg3 = b"To find, explore the funds offshore" + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipf: + with zipf.open('foo', mode='w') as w2: + w2.write(msg1) + with zipf.open('bar', mode='w') as w1: + with self.assertRaises(ValueError): + zipf.open('handle', mode='w') + with self.assertRaises(ValueError): + zipf.open('foo', mode='r') + with self.assertRaises(ValueError): + zipf.writestr('str', 'abcde') + with self.assertRaises(ValueError): + zipf.write(__file__, 'file') + with self.assertRaises(ValueError): + zipf.close() + w1.write(msg2) + with zipf.open('baz', mode='w') as w2: + w2.write(msg3) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipf: + self.assertEqual(zipf.read('foo'), msg1) + self.assertEqual(zipf.read('bar'), msg2) + self.assertEqual(zipf.read('baz'), msg3) + self.assertEqual(zipf.namelist(), ['foo', 'bar', 'baz']) + + def test_seek_tell(self): + # Test seek functionality + txt = b"Where's Bruce?" + bloc = txt.find(b"Bruce") + # Check seek on a file + with zipfile2.ZipFile(TESTFN, "w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(TESTFN, "r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + # Check seek on memory file + data = io.BytesIO() + with zipfile2.ZipFile(data, mode="w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(data, mode="r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + + @requires_bz2 + def test_decompress_without_3rd_party_library(self): + data = b'PK\x05\x06\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + zip_file = io.BytesIO(data) + with zipfile2.ZipFile(zip_file, 'w', compression=zipfile.ZIP_BZIP2) as zf: + zf.writestr('a.txt', b'a') + with mock.patch('zipfile.bz2', None): + with zipfile2.ZipFile(zip_file) as zf: + self.assertRaises(RuntimeError, zf.extract, 'a.txt') + + @requires_zlib + def test_full_overlap(self): + data = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e' + b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00a\xed' + b'\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\d\x0b`P' + b'K\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2' + b'\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00aPK' + b'\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e' + b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00bPK\x05' + b'\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00\x00/\x00\x00' + b'\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertEqual(zipf.namelist(), ['a', 'b']) + zi = zipf.getinfo('a') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + zi = zipf.getinfo('b') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + self.assertEqual(len(zipf.read('a')), 1033) + with self.assertRaisesRegex(zipfile.BadZipFile, 'File name.*differ'): + zipf.read('b') + + @requires_zlib + @unittest.skipIf(sys.version_info < (3, 8, 19), "Behaviour not supported in Python < 3.8.19") + def test_quoted_overlap(self): + data = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05Y\xfc' + b'8\x044\x00\x00\x00(\x04\x00\x00\x01\x00\x00\x00a\x00' + b'\x1f\x00\xe0\xffPK\x03\x04\x14\x00\x00\x00\x08\x00\xa0l' + b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00' + b'\x00\x00b\xed\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\' + b'd\x0b`PK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0' + b'lH\x05Y\xfc8\x044\x00\x00\x00(\x04\x00\x00\x01' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x00aPK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0l' + b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00$\x00\x00\x00' + b'bPK\x05\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00' + b'\x00S\x00\x00\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertEqual(zipf.namelist(), ['a', 'b']) + zi = zipf.getinfo('a') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 52) + self.assertEqual(zi.file_size, 1064) + zi = zipf.getinfo('b') + self.assertEqual(zi.header_offset, 36) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + with self.assertRaisesRegex(zipfile.BadZipFile, 'Overlapped entries'): + zipf.read('a') + self.assertEqual(len(zipf.read('b')), 1033) + + def tearDown(self): + unlink(TESTFN) + unlink(TESTFN2) + + +class AbstractBadCrcTests: + def test_testzip_with_bad_crc(self): + """Tests that files with bad CRCs return their name from testzip.""" + zipdata = self.zip_with_bad_crc + + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertEqual('afile', zipf.testzip()) + + def test_read_with_bad_crc(self): + """Tests that files with bad CRCs raise a BadZipFile exception when read.""" + zipdata = self.zip_with_bad_crc + + # Using ZipFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + self.assertRaises(zipfile.BadZipFile, zipf.read, 'afile') + + # Using ZipExtFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + self.assertRaises(zipfile.BadZipFile, corrupt_file.read) + + # Same with small reads (in order to exercise the buffering logic) + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + corrupt_file.MIN_READ_SIZE = 2 + with self.assertRaises(zipfile.BadZipFile): + while corrupt_file.read(2): + pass + + +class StoredBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_STORED + zip_with_bad_crc = ( + b'PK\003\004\024\0\0\0\0\0 \213\212;:r' + b'\253\377\f\0\0\0\f\0\0\0\005\0\0\000af' + b'ilehello,AworldP' + b'K\001\002\024\003\024\0\0\0\0\0 \213\212;:' + b'r\253\377\f\0\0\0\f\0\0\0\005\0\0\0\0' + b'\0\0\0\0\0\0\0\200\001\0\0\0\000afi' + b'lePK\005\006\0\0\0\0\001\0\001\0003\000' + b'\0\0/\0\0\0\0\0') + +@requires_zlib +class DeflateBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_DEFLATED + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00n}\x0c=FA' + b'KE\x10\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\xcbH\xcd\xc9\xc9W(\xcf/\xcaI\xc9\xa0' + b'=\x13\x00PK\x01\x02\x14\x03\x14\x00\x00\x00\x08\x00n' + b'}\x0c=FAKE\x10\x00\x00\x00n\x00\x00\x00\x05' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00' + b'\x00afilePK\x05\x06\x00\x00\x00\x00\x01\x00' + b'\x01\x003\x00\x00\x003\x00\x00\x00\x00\x00') + +@requires_bz2 +class Bzip2BadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_BZIP2 + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0c\x00nu\x0c=FA' + b'KE8\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ileBZh91AY&SY\xd4\xa8\xca' + b'\x7f\x00\x00\x0f\x11\x80@\x00\x06D\x90\x80 \x00 \xa5' + b'P\xd9!\x03\x03\x13\x13\x13\x89\xa9\xa9\xc2u5:\x9f' + b'\x8b\xb9"\x9c(HjTe?\x80PK\x01\x02\x14' + b'\x03\x14\x03\x00\x00\x0c\x00nu\x0c=FAKE8' + b'\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x00 \x80\x80\x81\x00\x00\x00\x00afilePK' + b'\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00\x00[\x00' + b'\x00\x00\x00\x00') + +@requires_lzma +class LzmaBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_LZMA + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\t\x04\x05\x00]\x00\x00\x00\x04\x004\x19I' + b'\xee\x8d\xe9\x17\x89:3`\tq!.8\x00PK' + b'\x01\x02\x14\x03\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00 \x80\x80\x81\x00\x00\x00\x00afil' + b'ePK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00' + b'\x00>\x00\x00\x00\x00\x00') + + +class DecryptionTests(unittest.TestCase): + """Check that ZIP decryption works. Since the library does not + support encryption at the moment, we use a pre-generated encrypted + ZIP file.""" + + data = ( + b'PK\x03\x04\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00\x1a\x00' + b'\x00\x00\x08\x00\x00\x00test.txt\xfa\x10\xa0gly|\xfa-\xc5\xc0=\xf9y' + b'\x18\xe0\xa8r\xb3Z}Lg\xbc\xae\xf9|\x9b\x19\xe4\x8b\xba\xbb)\x8c\xb0\xdbl' + b'PK\x01\x02\x14\x00\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00' + b'\x1a\x00\x00\x00\x08\x00\x00\x00\x00\x00\x00\x00\x01\x00 \x00\xb6\x81' + b'\x00\x00\x00\x00test.txtPK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x006\x00' + b'\x00\x00L\x00\x00\x00\x00\x00' ) + data2 = ( + b'PK\x03\x04\x14\x00\t\x00\x08\x00\xcf}38xu\xaa\xb2\x14\x00\x00\x00\x00\x02' + b'\x00\x00\x04\x00\x15\x00zeroUT\t\x00\x03\xd6\x8b\x92G\xda\x8b\x92GUx\x04' + b'\x00\xe8\x03\xe8\x03\xc7= 3.8.20") + def test_malformed_paths(self): + """ + Path should handle malformed paths gracefully. + + Paths with leading slashes are not visible. + + Paths with dots are treated like regular files. + """ + data = io.BytesIO() + zf = zipfile2.ZipFile(data, "w") + zf.writestr("/one-slash.txt", b"content") + zf.writestr("//two-slash.txt", b"content") + zf.writestr("../parent.txt", b"content") + zf.filename = '' + root = zipfile.Path(zf) + assert list(map(str, root.iterdir())) == ['../'] + assert root.joinpath('..').joinpath('parent.txt').read_bytes() == b'content' + + def test_unsupported_names(self): + """ + Path segments with special characters are readable. + + On some platforms or file systems, characters like + ``:`` and ``?`` are not allowed, but they are valid + in the zip file. + """ + data = io.BytesIO() + zf = zipfile2.ZipFile(data, "w") + zf.writestr("path?", b"content") + zf.writestr("V: NMS.flac", b"fLaC...") + zf.filename = '' + root = zipfile.Path(zf) + contents = root.iterdir() + assert next(contents).name == 'path?' + assert next(contents).name == 'V: NMS.flac' + assert root.joinpath('V: NMS.flac').read_bytes() == b"fLaC..." + + def test_backslash_not_separator(self): + """ + In a zip file, backslashes are not separators. + """ + data = io.BytesIO() + zf = zipfile2.ZipFile(data, "w") + zf.writestr(DirtyZipInfo.for_name("foo\\bar", zf), b"content") + zf.filename = '' + root = zipfile.Path(zf) + (first,) = root.iterdir() + assert not first.is_dir() + assert first.name == 'foo\\bar' + + +class DirtyZipInfo(zipfile.ZipInfo): + """ + Bypass name sanitization. + """ + + def __init__(self, filename, *args, **kwargs): + super().__init__(filename, *args, **kwargs) + self.filename = filename + + @classmethod + def for_name(cls, name, archive): + """ + Construct the same way that ZipFile.writestr does. + + TODO: extract this functionality and re-use + """ + self = cls(filename=name, date_time=time.localtime(time.time())[:6]) + self.compress_type = archive.compression + self.compress_level = archive.compresslevel + if self.filename.endswith('/'): # pragma: no cover + self.external_attr = 0o40775 << 16 # drwxrwxr-x + self.external_attr |= 0x10 # MS-DOS directory flag + else: + self.external_attr = 0o600 << 16 # ?rw------- + return self + + +if __name__ == "__main__": + unittest.main() diff --git a/cpython-tests/3.8/test_zipfile64.py b/cpython-tests/3.8/test_zipfile64.py new file mode 100644 index 0000000..7b99aef --- /dev/null +++ b/cpython-tests/3.8/test_zipfile64.py @@ -0,0 +1,147 @@ +# Tests of the full ZIP64 functionality of zipfile +# The support.requires call is the only reason for keeping this separate +# from test_zipfile +from test import support + +# XXX(nnorwitz): disable this test by looking for extralargefile resource, +# which doesn't exist. This test takes over 30 minutes to run in general +# and requires more disk space than most of the buildbots. +support.requires( + 'extralargefile', + 'test requires loads of disk-space bytes and a long time to run' + ) + +import zipfile, os, unittest +import time +import sys + +from tempfile import TemporaryFile + +from test.support import TESTFN, requires_zlib + +import zipfile2 + +TESTFN2 = TESTFN + "2" + +# How much time in seconds can pass before we print a 'Still working' message. +_PRINT_WORKING_MSG_INTERVAL = 60 + +class TestsWithSourceFile(unittest.TestCase): + def setUp(self): + # Create test data. + line_gen = ("Test of zipfile line %d." % i for i in range(1000000)) + self.data = '\n'.join(line_gen).encode('ascii') + + # And write it to a file. + with open(TESTFN, "wb") as fp: + fp.write(self.data) + + def zipTest(self, f, compression): + # Create the ZIP archive. + with zipfile2.ZipFile(f, "w", compression) as zipfp: + + # It will contain enough copies of self.data to reach about 6 GiB of + # raw data to store. + filecount = 6*1024**3 // len(self.data) + + next_time = time.monotonic() + _PRINT_WORKING_MSG_INTERVAL + for num in range(filecount): + zipfp.writestr("testfn%d" % num, self.data) + # Print still working message since this test can be really slow + if next_time <= time.monotonic(): + next_time = time.monotonic() + _PRINT_WORKING_MSG_INTERVAL + print(( + ' zipTest still writing %d of %d, be patient...' % + (num, filecount)), file=sys.__stdout__) + sys.__stdout__.flush() + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + for num in range(filecount): + self.assertEqual(zipfp.read("testfn%d" % num), self.data) + # Print still working message since this test can be really slow + if next_time <= time.monotonic(): + next_time = time.monotonic() + _PRINT_WORKING_MSG_INTERVAL + print(( + ' zipTest still reading %d of %d, be patient...' % + (num, filecount)), file=sys.__stdout__) + sys.__stdout__.flush() + + def testStored(self): + # Try the temp file first. If we do TESTFN2 first, then it hogs + # gigabytes of disk space for the duration of the test. + with TemporaryFile() as f: + self.zipTest(f, zipfile.ZIP_STORED) + self.assertFalse(f.closed) + self.zipTest(TESTFN2, zipfile.ZIP_STORED) + + @requires_zlib + def testDeflated(self): + # Try the temp file first. If we do TESTFN2 first, then it hogs + # gigabytes of disk space for the duration of the test. + with TemporaryFile() as f: + self.zipTest(f, zipfile.ZIP_DEFLATED) + self.assertFalse(f.closed) + self.zipTest(TESTFN2, zipfile.ZIP_DEFLATED) + + def tearDown(self): + for fname in TESTFN, TESTFN2: + if os.path.exists(fname): + os.remove(fname) + + +class OtherTests(unittest.TestCase): + def testMoreThan64kFiles(self): + # This test checks that more than 64k files can be added to an archive, + # and that the resulting archive can be read properly by ZipFile + with zipfile2.ZipFile(TESTFN, mode="w", allowZip64=True) as zipf: + zipf.debug = 100 + numfiles = (1 << 16) * 3//2 + for i in range(numfiles): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles) + + with zipfile2.ZipFile(TESTFN, mode="r") as zipf2: + self.assertEqual(len(zipf2.namelist()), numfiles) + for i in range(numfiles): + content = zipf2.read("foo%08d" % i).decode('ascii') + self.assertEqual(content, "%d" % (i**3 % 57)) + + def testMoreThan64kFilesAppend(self): + with zipfile2.ZipFile(TESTFN, mode="w", allowZip64=False) as zipf: + zipf.debug = 100 + numfiles = (1 << 16) - 1 + for i in range(numfiles): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles) + with self.assertRaises(zipfile.LargeZipFile): + zipf.writestr("foo%08d" % numfiles, b'') + self.assertEqual(len(zipf.namelist()), numfiles) + + with zipfile2.ZipFile(TESTFN, mode="a", allowZip64=False) as zipf: + zipf.debug = 100 + self.assertEqual(len(zipf.namelist()), numfiles) + with self.assertRaises(zipfile.LargeZipFile): + zipf.writestr("foo%08d" % numfiles, b'') + self.assertEqual(len(zipf.namelist()), numfiles) + + with zipfile2.ZipFile(TESTFN, mode="a", allowZip64=True) as zipf: + zipf.debug = 100 + self.assertEqual(len(zipf.namelist()), numfiles) + numfiles2 = (1 << 16) * 3//2 + for i in range(numfiles, numfiles2): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles2) + + with zipfile2.ZipFile(TESTFN, mode="r") as zipf2: + self.assertEqual(len(zipf2.namelist()), numfiles2) + for i in range(numfiles2): + content = zipf2.read("foo%08d" % i).decode('ascii') + self.assertEqual(content, "%d" % (i**3 % 57)) + + def tearDown(self): + support.unlink(TESTFN) + support.unlink(TESTFN2) + +if __name__ == "__main__": + unittest.main() diff --git a/cpython-tests/3.8/test_zipimport.py b/cpython-tests/3.8/test_zipimport.py new file mode 100644 index 0000000..f628b8f --- /dev/null +++ b/cpython-tests/3.8/test_zipimport.py @@ -0,0 +1,780 @@ +import sys +import os +import marshal +import importlib +import importlib.util +import struct +import time +import unittest +import unittest.mock + +from test import support + +from zipfile import ZipInfo, ZIP_STORED, ZIP_DEFLATED + +import zipimport +import linecache +import doctest +import inspect +import io +from traceback import extract_tb, extract_stack, print_tb +try: + import zlib +except ImportError: + zlib = None + +from zipfile2 import ZipFile + + +test_src = """\ +def get_name(): + return __name__ +def get_file(): + return __file__ +""" +test_co = compile(test_src, "", "exec") +raise_src = 'def do_raise(): raise TypeError\n' + +def make_pyc(co, mtime, size): + data = marshal.dumps(co) + if type(mtime) is type(0.0): + # Mac mtimes need a bit of special casing + if mtime < 0x7fffffff: + mtime = int(mtime) + else: + mtime = int(-0x100000000 + int(mtime)) + pyc = (importlib.util.MAGIC_NUMBER + + struct.pack("", "exec"), NOW, len(src)) + files = {TESTMOD + pyc_ext: (NOW, pyc), + "some.data": (NOW, "some data")} + self.doTest(pyc_ext, files, TESTMOD) + + def testDefaultOptimizationLevel(self): + # zipimport should use the default optimization level (#28131) + src = """if 1: # indent hack + def test(val): + assert(val) + return val\n""" + files = {TESTMOD + '.py': (NOW, src)} + self.makeZip(files) + sys.path.insert(0, TEMP_ZIP) + mod = importlib.import_module(TESTMOD) + self.assertEqual(mod.test(1), 1) + self.assertRaises(AssertionError, mod.test, False) + + def testImport_WithStuff(self): + # try importing from a zipfile which contains additional + # stuff at the beginning of the file + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, + stuff=b"Some Stuff"*31) + + def assertModuleSource(self, module): + self.assertEqual(inspect.getsource(module), test_src) + + def testGetSource(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, call=self.assertModuleSource) + + def testGetCompiledSource(self): + pyc = make_pyc(compile(test_src, "", "exec"), NOW, len(test_src)) + files = {TESTMOD + ".py": (NOW, test_src), + TESTMOD + pyc_ext: (NOW, pyc)} + self.doTest(pyc_ext, files, TESTMOD, call=self.assertModuleSource) + + def runDoctest(self, callback): + files = {TESTMOD + ".py": (NOW, test_src), + "xyz.txt": (NOW, ">>> log.append(True)\n")} + self.doTest(".py", files, TESTMOD, call=callback) + + def doDoctestFile(self, module): + log = [] + old_master, doctest.master = doctest.master, None + try: + doctest.testfile( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ) + finally: + doctest.master = old_master + self.assertEqual(log,[True]) + + def testDoctestFile(self): + self.runDoctest(self.doDoctestFile) + + def doDoctestSuite(self, module): + log = [] + doctest.DocFileTest( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ).run() + self.assertEqual(log,[True]) + + def testDoctestSuite(self): + self.runDoctest(self.doDoctestSuite) + + def doTraceback(self, module): + try: + module.do_raise() + except: + tb = sys.exc_info()[2].tb_next + + f,lno,n,line = extract_tb(tb, 1)[0] + self.assertEqual(line, raise_src.strip()) + + f,lno,n,line = extract_stack(tb.tb_frame, 1)[0] + self.assertEqual(line, raise_src.strip()) + + s = io.StringIO() + print_tb(tb, 1, s) + self.assertTrue(s.getvalue().endswith(raise_src)) + else: + raise AssertionError("This ought to be impossible") + + def testTraceback(self): + files = {TESTMOD + ".py": (NOW, raise_src)} + self.doTest(None, files, TESTMOD, call=self.doTraceback) + + @unittest.skipIf(support.TESTFN_UNENCODABLE is None, + "need an unencodable filename") + def testUnencodable(self): + filename = support.TESTFN_UNENCODABLE + ".zip" + self.addCleanup(support.unlink, filename) + with ZipFile(filename, "w") as z: + zinfo = ZipInfo(TESTMOD + ".py", time.localtime(NOW)) + zinfo.compress_type = self.compression + z.writestr(zinfo, test_src) + zipimport.zipimporter(filename).load_module(TESTMOD) + + def testBytesPath(self): + filename = support.TESTFN + ".zip" + self.addCleanup(support.unlink, filename) + with ZipFile(filename, "w") as z: + zinfo = ZipInfo(TESTMOD + ".py", time.localtime(NOW)) + zinfo.compress_type = self.compression + z.writestr(zinfo, test_src) + + zipimport.zipimporter(filename) + zipimport.zipimporter(os.fsencode(filename)) + with self.assertRaises(TypeError): + zipimport.zipimporter(bytearray(os.fsencode(filename))) + with self.assertRaises(TypeError): + zipimport.zipimporter(memoryview(os.fsencode(filename))) + + def testComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, comment=b"comment") + + def testBeginningCruftAndComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, stuff=b"cruft" * 64, comment=b"hi") + + def testLargestPossibleComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, comment=b"c" * ((1 << 16) - 1)) + + +@support.requires_zlib +class CompressedZipImportTestCase(UncompressedZipImportTestCase): + compression = ZIP_DEFLATED + + +class BadFileZipImportTestCase(unittest.TestCase): + def assertZipFailure(self, filename): + self.assertRaises(zipimport.ZipImportError, + zipimport.zipimporter, filename) + + def testNoFile(self): + self.assertZipFailure('AdfjdkFJKDFJjdklfjs') + + def testEmptyFilename(self): + self.assertZipFailure('') + + def testBadArgs(self): + self.assertRaises(TypeError, zipimport.zipimporter, None) + self.assertRaises(TypeError, zipimport.zipimporter, TESTMOD, kwd=None) + self.assertRaises(TypeError, zipimport.zipimporter, + list(os.fsencode(TESTMOD))) + + def testFilenameTooLong(self): + self.assertZipFailure('A' * 33000) + + def testEmptyFile(self): + support.unlink(TESTMOD) + support.create_empty_file(TESTMOD) + self.assertZipFailure(TESTMOD) + + def testFileUnreadable(self): + support.unlink(TESTMOD) + fd = os.open(TESTMOD, os.O_CREAT, 000) + try: + os.close(fd) + + with self.assertRaises(zipimport.ZipImportError) as cm: + zipimport.zipimporter(TESTMOD) + finally: + # If we leave "the read-only bit" set on Windows, nothing can + # delete TESTMOD, and later tests suffer bogus failures. + os.chmod(TESTMOD, 0o666) + support.unlink(TESTMOD) + + def testNotZipFile(self): + support.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write('a' * 22) + fp.close() + self.assertZipFailure(TESTMOD) + + # XXX: disabled until this works on Big-endian machines + def _testBogusZipFile(self): + support.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write(struct.pack('=I', 0x06054B50)) + fp.write('a' * 18) + fp.close() + z = zipimport.zipimporter(TESTMOD) + + try: + self.assertRaises(TypeError, z.find_module, None) + self.assertRaises(TypeError, z.load_module, None) + self.assertRaises(TypeError, z.is_package, None) + self.assertRaises(TypeError, z.get_code, None) + self.assertRaises(TypeError, z.get_data, None) + self.assertRaises(TypeError, z.get_source, None) + + error = zipimport.ZipImportError + self.assertEqual(z.find_module('abc'), None) + + self.assertRaises(error, z.load_module, 'abc') + self.assertRaises(error, z.get_code, 'abc') + self.assertRaises(OSError, z.get_data, 'abc') + self.assertRaises(error, z.get_source, 'abc') + self.assertRaises(error, z.is_package, 'abc') + finally: + zipimport._zip_directory_cache.clear() + + +def test_main(): + try: + support.run_unittest( + UncompressedZipImportTestCase, + CompressedZipImportTestCase, + BadFileZipImportTestCase, + ) + finally: + support.unlink(TESTMOD) + +if __name__ == "__main__": + test_main() diff --git a/cpython-tests/3.8/zip_cp437_header.zip b/cpython-tests/3.8/zip_cp437_header.zip new file mode 100644 index 0000000..f7c6cf1 Binary files /dev/null and b/cpython-tests/3.8/zip_cp437_header.zip differ diff --git a/cpython-tests/3.8/zipdir.zip b/cpython-tests/3.8/zipdir.zip new file mode 100644 index 0000000..ac21d7a Binary files /dev/null and b/cpython-tests/3.8/zipdir.zip differ diff --git a/cpython-tests/3.8/ziptestdata/README.md b/cpython-tests/3.8/ziptestdata/README.md new file mode 100644 index 0000000..6b9147d --- /dev/null +++ b/cpython-tests/3.8/ziptestdata/README.md @@ -0,0 +1,35 @@ +# Test data for `test_zipfile` + +The test executables in this directory are created manually from header.sh and +the `testdata_module_inside_zip.py` file. You must have infozip's zip utility +installed (`apt install zip` on Debian). + +## Purpose + +These are used to test executable files with an appended zipfile, in a scenario +where the executable is _not_ a Python interpreter itself so our automatic +zipimport machinery (that'd look for `__main__.py`) is not being used. + +## Updating the test executables + +If you update header.sh or the testdata_module_inside_zip.py file, rerun the +commands below. These are expected to be rarely changed, if ever. + +### Standard old format (2.0) zip file + +``` +zip -0 zip2.zip testdata_module_inside_zip.py +cat header.sh zip2.zip >exe_with_zip +rm zip2.zip +``` + +### Modern format (4.5) zip64 file + +Redirecting from stdin forces infozip's zip tool to create a zip64. + +``` +zip -0 zip64.zip +cat header.sh zip64.zip >exe_with_z64 +rm zip64.zip +``` + diff --git a/cpython-tests/3.8/ziptestdata/exe_with_z64 b/cpython-tests/3.8/ziptestdata/exe_with_z64 new file mode 100644 index 0000000..82b03cf Binary files /dev/null and b/cpython-tests/3.8/ziptestdata/exe_with_z64 differ diff --git a/cpython-tests/3.8/ziptestdata/exe_with_zip b/cpython-tests/3.8/ziptestdata/exe_with_zip new file mode 100644 index 0000000..c833cdf Binary files /dev/null and b/cpython-tests/3.8/ziptestdata/exe_with_zip differ diff --git a/cpython-tests/3.8/ziptestdata/header.sh b/cpython-tests/3.8/ziptestdata/header.sh new file mode 100644 index 0000000..52dc91a --- /dev/null +++ b/cpython-tests/3.8/ziptestdata/header.sh @@ -0,0 +1,24 @@ +#!/bin/bash +INTERPRETER_UNDER_TEST="$1" +if [[ ! -x "${INTERPRETER_UNDER_TEST}" ]]; then + echo "Interpreter must be the command line argument." + exit 4 +fi +EXECUTABLE="$0" exec "${INTERPRETER_UNDER_TEST}" -E - <e|f"g?h*i', ','), r'a\b,c_d_e_f_g_h_i') + self.assertEqual(san('../../foo../../ba..r', '/'), r'foo/ba..r') + + def test_extract_hackers_arcnames_common_cases(self): + common_hacknames = [ + ('../foo/bar', 'foo/bar'), + ('foo/../bar', 'foo/bar'), + ('foo/../../bar', 'foo/bar'), + ('foo/bar/..', 'foo/bar'), + ('./../foo/bar', 'foo/bar'), + ('/foo/bar', 'foo/bar'), + ('/foo/../bar', 'foo/bar'), + ('/foo/../../bar', 'foo/bar'), + ] + self._test_extract_hackers_arcnames(common_hacknames) + + @unittest.skipIf(os.path.sep != '\\', 'Requires \\ as path separator.') + def test_extract_hackers_arcnames_windows_only(self): + """Test combination of path fixing and windows name sanitization.""" + windows_hacknames = [ + (r'..\foo\bar', 'foo/bar'), + (r'..\/foo\/bar', 'foo/bar'), + (r'foo/\..\/bar', 'foo/bar'), + (r'foo\/../\bar', 'foo/bar'), + (r'C:foo/bar', 'foo/bar'), + (r'C:/foo/bar', 'foo/bar'), + (r'C://foo/bar', 'foo/bar'), + (r'C:\foo\bar', 'foo/bar'), + (r'//conky/mountpoint/foo/bar', 'foo/bar'), + (r'\\conky\mountpoint\foo\bar', 'foo/bar'), + (r'///conky/mountpoint/foo/bar', 'conky/mountpoint/foo/bar'), + (r'\\\conky\mountpoint\foo\bar', 'conky/mountpoint/foo/bar'), + (r'//conky//mountpoint/foo/bar', 'conky/mountpoint/foo/bar'), + (r'\\conky\\mountpoint\foo\bar', 'conky/mountpoint/foo/bar'), + (r'//?/C:/foo/bar', 'foo/bar'), + (r'\\?\C:\foo\bar', 'foo/bar'), + (r'C:/../C:/foo/bar', 'C_/foo/bar'), + (r'a:b\ce|f"g?h*i', 'b/c_d_e_f_g_h_i'), + ('../../foo../../ba..r', 'foo/ba..r'), + ] + self._test_extract_hackers_arcnames(windows_hacknames) + + @unittest.skipIf(os.path.sep != '/', r'Requires / as path separator.') + def test_extract_hackers_arcnames_posix_only(self): + posix_hacknames = [ + ('//foo/bar', 'foo/bar'), + ('../../foo../../ba..r', 'foo../ba..r'), + (r'foo/..\bar', r'foo/..\bar'), + ] + self._test_extract_hackers_arcnames(posix_hacknames) + + def _test_extract_hackers_arcnames(self, hacknames): + for arcname, fixedname in hacknames: + content = b'foobar' + arcname.encode() + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipfp: + zinfo = zipfile.ZipInfo() + # preserve backslashes + zinfo.filename = arcname + zinfo.external_attr = 0o600 << 16 + zipfp.writestr(zinfo, content) + + arcname = arcname.replace(os.sep, "/") + targetpath = os.path.join('target', 'subdir', 'subsub') + correctfile = os.path.join(targetpath, *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname, targetpath) + self.assertEqual(writtenfile, correctfile, + msg='extract %r: %r != %r' % + (arcname, writtenfile, correctfile)) + self.check_file(correctfile, content) + rmtree('target') + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall(targetpath) + self.check_file(correctfile, content) + rmtree('target') + + correctfile = os.path.join(os.getcwd(), *fixedname.split('/')) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + writtenfile = zipfp.extract(arcname) + self.assertEqual(writtenfile, correctfile, + msg="extract %r" % arcname) + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipfp: + zipfp.extractall() + self.check_file(correctfile, content) + rmtree(fixedname.split('/')[0]) + + unlink(TESTFN2) + + +class OtherTests(unittest.TestCase): + + def test_open_via_zip_info(self): + # Create the ZIP archive + with zipfile2.ZipFile(TESTFN2, "w", zipfile.ZIP_STORED, low_level=True) as zipfp: + zipfp.writestr("name", "foo") + with self.assertWarns(UserWarning): + zipfp.writestr("name", "bar") + self.assertEqual(zipfp.namelist(), ["name"] * 2) + + with zipfile2.ZipFile(TESTFN2, "r", low_level=True) as zipfp: + infos = zipfp.infolist() + data = b"" + for info in infos: + with zipfp.open(info) as zipopen: + data += zipopen.read() + self.assertIn(data, {b"foobar", b"barfoo"}) + data = b"" + for info in infos: + data += zipfp.read(info) + self.assertIn(data, {b"foobar", b"barfoo"}) + + def test_writestr_extended_local_header_issue1202(self): + with zipfile2.ZipFile(TESTFN2, 'w') as orig_zip: + for data in 'abcdefghijklmnop': + zinfo = zipfile.ZipInfo(data) + zinfo.flag_bits |= 0x08 # Include an extended local header. + orig_zip.writestr(zinfo, data) + + def test_close(self): + """Check that the zipfile is closed after the 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertIsNotNone(zipfp.fp, 'zipfp is not open') + self.assertIsNone(zipfp.fp, 'zipfp is not closed') + + def test_close_on_exception(self): + """Check that the zipfile is closed if an exception is raised in the + 'with' block.""" + with zipfile2.ZipFile(TESTFN2, "w") as zipfp: + for fpath, fdata in SMALL_TEST_DATA: + zipfp.writestr(fpath, fdata) + + try: + with zipfile2.ZipFile(TESTFN2, "r") as zipfp2: + raise zipfile.BadZipFile() + except zipfile.BadZipFile: + self.assertIsNone(zipfp2.fp, 'zipfp is not closed') + + def test_unsupported_version(self): + # File has an extract_version of 120 + data = (b'PK\x03\x04x\x00\x00\x00\x00\x00!p\xa1@\x00\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x01\x00\x00\x00xPK\x01\x02x\x03x\x00\x00\x00\x00' + b'\x00!p\xa1@\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x01\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00\x00xPK\x05\x06' + b'\x00\x00\x00\x00\x01\x00\x01\x00/\x00\x00\x00\x1f\x00\x00\x00\x00\x00') + + self.assertRaises(NotImplementedError, zipfile2.ZipFile, + io.BytesIO(data), 'r') + + @requires_zlib() + def test_read_unicode_filenames(self): + # bug #10801 + fname = findfile('zip_cp437_header.zip') + with zipfile2.ZipFile(fname) as zipfp: + for name in zipfp.namelist(): + zipfp.open(name).close() + + def test_write_unicode_filenames(self): + with zipfile2.ZipFile(TESTFN, "w") as zf: + zf.writestr("foo.txt", "Test for unicode filename") + zf.writestr("\xf6.txt", "Test for unicode filename") + self.assertIsInstance(zf.infolist()[0].filename, str) + + with zipfile2.ZipFile(TESTFN, "r") as zf: + self.assertEqual(zf.filelist[0].filename, "foo.txt") + self.assertEqual(zf.filelist[1].filename, "\xf6.txt") + + def test_read_after_write_unicode_filenames(self): + with zipfile2.ZipFile(TESTFN2, 'w') as zipfp: + zipfp.writestr('приклад', b'sample') + self.assertEqual(zipfp.read('приклад'), b'sample') + + def test_exclusive_create_zip_file(self): + """Test exclusive creating a new zipfile.""" + unlink(TESTFN2) + filename = 'testfile.txt' + content = b'hello, world. this is some content.' + with zipfile2.ZipFile(TESTFN2, "x", zipfile.ZIP_STORED) as zipfp: + zipfp.writestr(filename, content) + with self.assertRaises(FileExistsError): + zipfile2.ZipFile(TESTFN2, "x", zipfile.ZIP_STORED) + with zipfile2.ZipFile(TESTFN2, "r") as zipfp: + self.assertEqual(zipfp.namelist(), [filename]) + self.assertEqual(zipfp.read(filename), content) + + def test_create_non_existent_file_for_append(self): + if os.path.exists(TESTFN): + os.unlink(TESTFN) + + filename = 'testfile.txt' + content = b'hello, world. this is some content.' + + try: + with zipfile2.ZipFile(TESTFN, 'a') as zf: + zf.writestr(filename, content) + except OSError: + self.fail('Could not append data to a non-existent zip file.') + + self.assertTrue(os.path.exists(TESTFN)) + + with zipfile2.ZipFile(TESTFN, 'r') as zf: + self.assertEqual(zf.read(filename), content) + + def test_close_erroneous_file(self): + # This test checks that the ZipFile constructor closes the file object + # it opens if there's an error in the file. If it doesn't, the + # traceback holds a reference to the ZipFile object and, indirectly, + # the file object. + # On Windows, this causes the os.unlink() call to fail because the + # underlying file is still open. This is SF bug #412214. + # + with open(TESTFN, "w") as fp: + fp.write("this is not a legal zip file\n") + try: + zf = zipfile2.ZipFile(TESTFN) + except zipfile.BadZipFile: + pass + + def test_is_zip_erroneous_file(self): + """Check that is_zipfile() correctly identifies non-zip files.""" + # - passing a filename + with open(TESTFN, "w") as fp: + fp.write("this is not a legal zip file\n") + self.assertFalse(zipfile.is_zipfile(TESTFN)) + # - passing a path-like object + self.assertFalse(zipfile.is_zipfile(pathlib.Path(TESTFN))) + # - passing a file object + with open(TESTFN, "rb") as fp: + self.assertFalse(zipfile.is_zipfile(fp)) + # - passing a file-like object + fp = io.BytesIO() + fp.write(b"this is not a legal zip file\n") + self.assertFalse(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + self.assertFalse(zipfile.is_zipfile(fp)) + + def test_damaged_zipfile(self): + """Check that zipfiles with missing bytes at the end raise BadZipFile.""" + # - Create a valid zip file + fp = io.BytesIO() + with zipfile2.ZipFile(fp, mode="w") as zipf: + zipf.writestr("foo.txt", b"O, for a Muse of Fire!") + zipfiledata = fp.getvalue() + + # - Now create copies of it missing the last N bytes and make sure + # a BadZipFile exception is raised when we try to open it + for N in range(len(zipfiledata)): + fp = io.BytesIO(zipfiledata[:N]) + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, fp) + + def test_is_zip_valid_file(self): + """Check that is_zipfile() correctly identifies zip files.""" + # - passing a filename + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt", b"O, for a Muse of Fire!") + + self.assertTrue(zipfile.is_zipfile(TESTFN)) + # - passing a file object + with open(TESTFN, "rb") as fp: + self.assertTrue(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + zip_contents = fp.read() + # - passing a file-like object + fp = io.BytesIO() + fp.write(zip_contents) + self.assertTrue(zipfile.is_zipfile(fp)) + fp.seek(0, 0) + self.assertTrue(zipfile.is_zipfile(fp)) + + def test_non_existent_file_raises_OSError(self): + # make sure we don't raise an AttributeError when a partially-constructed + # ZipFile instance is finalized; this tests for regression on SF tracker + # bug #403871. + + # The bug we're testing for caused an AttributeError to be raised + # when a ZipFile instance was created for a file that did not + # exist; the .fp member was not initialized but was needed by the + # __del__() method. Since the AttributeError is in the __del__(), + # it is ignored, but the user should be sufficiently annoyed by + # the message on the output that regression will be noticed + # quickly. + self.assertRaises(OSError, zipfile2.ZipFile, TESTFN) + + def test_empty_file_raises_BadZipFile(self): + f = open(TESTFN, 'w') + f.close() + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN) + + with open(TESTFN, 'w') as fp: + fp.write("short file") + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN) + + def test_closed_zip_raises_ValueError(self): + """Verify that testzip() doesn't swallow inappropriate exceptions.""" + data = io.BytesIO() + with zipfile2.ZipFile(data, mode="w") as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + + # This is correct; calling .read on a closed ZipFile should raise + # a ValueError, and so should calling .testzip. An earlier + # version of .testzip would swallow this exception (and any other) + # and report that the first file in the archive was corrupt. + self.assertRaises(ValueError, zipf.read, "foo.txt") + self.assertRaises(ValueError, zipf.open, "foo.txt") + self.assertRaises(ValueError, zipf.testzip) + self.assertRaises(ValueError, zipf.writestr, "bogus.txt", "bogus") + with open(TESTFN, 'w') as f: + f.write('zipfile test data') + self.assertRaises(ValueError, zipf.write, TESTFN) + + def test_bad_constructor_mode(self): + """Check that bad modes passed to ZipFile constructor are caught.""" + self.assertRaises(ValueError, zipfile2.ZipFile, TESTFN, "q") + + def test_bad_open_mode(self): + """Check that bad modes passed to ZipFile.open are caught.""" + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + + with zipfile2.ZipFile(TESTFN, mode="r") as zipf: + # read the data to make sure the file is there + zipf.read("foo.txt") + self.assertRaises(ValueError, zipf.open, "foo.txt", "q") + # universal newlines support is removed + self.assertRaises(ValueError, zipf.open, "foo.txt", "U") + self.assertRaises(ValueError, zipf.open, "foo.txt", "rU") + + def test_read0(self): + """Check that calling read(0) on a ZipExtFile object returns an empty + string and doesn't advance file pointer.""" + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + # read the data to make sure the file is there + with zipf.open("foo.txt") as f: + for i in range(FIXEDTEST_SIZE): + self.assertEqual(f.read(0), b'') + + self.assertEqual(f.read(), b"O, for a Muse of Fire!") + + def test_open_non_existent_item(self): + """Check that attempting to call open() for an item that doesn't + exist in the archive raises a RuntimeError.""" + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + self.assertRaises(KeyError, zipf.open, "foo.txt", "r") + + def test_bad_compression_mode(self): + """Check that bad compression methods passed to ZipFile.open are + caught.""" + self.assertRaises(NotImplementedError, zipfile2.ZipFile, TESTFN, "w", -1) + + def test_unsupported_compression(self): + # data is declared as shrunk, but actually deflated + data = (b'PK\x03\x04.\x00\x00\x00\x01\x00\xe4C\xa1@\x00\x00\x00' + b'\x00\x02\x00\x00\x00\x00\x00\x00\x00\x01\x00\x00\x00x\x03\x00PK\x01' + b'\x02.\x03.\x00\x00\x00\x01\x00\xe4C\xa1@\x00\x00\x00\x00\x02\x00\x00' + b'\x00\x00\x00\x00\x00\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x80\x01\x00\x00\x00\x00xPK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x00' + b'/\x00\x00\x00!\x00\x00\x00\x00\x00') + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertRaises(NotImplementedError, zipf.open, 'x') + + def test_null_byte_in_filename(self): + """Check that a filename containing a null byte is properly + terminated.""" + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.writestr("foo.txt\x00qqq", b"O, for a Muse of Fire!") + self.assertEqual(zipf.namelist(), ['foo.txt']) + + def test_struct_sizes(self): + """Check that ZIP internal structure sizes are calculated correctly.""" + self.assertEqual(zipfile.sizeEndCentDir, 22) + self.assertEqual(zipfile.sizeCentralDir, 46) + self.assertEqual(zipfile.sizeEndCentDir64, 56) + self.assertEqual(zipfile.sizeEndCentDir64Locator, 20) + + def test_comments(self): + """Check that comments on the archive are handled properly.""" + + # check default comment is empty + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + self.assertEqual(zipf.comment, b'') + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + + with zipfile2.ZipFile(TESTFN, mode="r") as zipfr: + self.assertEqual(zipfr.comment, b'') + + # check a simple short comment + comment = b'Bravely taking to his feet, he beat a very brave retreat.' + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.comment = comment + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN, mode="r") as zipfr: + self.assertEqual(zipf.comment, comment) + + # check a comment of max length + comment2 = ''.join(['%d' % (i**3 % 10) for i in range((1 << 16)-1)]) + comment2 = comment2.encode("ascii") + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + zipf.comment = comment2 + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + + with zipfile2.ZipFile(TESTFN, mode="r") as zipfr: + self.assertEqual(zipfr.comment, comment2) + + # check a comment that is too long is truncated + with zipfile2.ZipFile(TESTFN, mode="w") as zipf: + with self.assertWarns(UserWarning): + zipf.comment = comment2 + b'oops' + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN, mode="r") as zipfr: + self.assertEqual(zipfr.comment, comment2) + + # check that comments are correctly modified in append mode + with zipfile2.ZipFile(TESTFN,mode="w") as zipf: + zipf.comment = b"original comment" + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN,mode="a") as zipf: + zipf.comment = b"an updated comment" + with zipfile2.ZipFile(TESTFN,mode="r") as zipf: + self.assertEqual(zipf.comment, b"an updated comment") + + # check that comments are correctly shortened in append mode + # and the file is indeed truncated + with zipfile2.ZipFile(TESTFN,mode="w") as zipf: + zipf.comment = b"original comment that's longer" + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + original_zip_size = os.path.getsize(TESTFN) + with zipfile2.ZipFile(TESTFN,mode="a") as zipf: + zipf.comment = b"shorter comment" + self.assertTrue(original_zip_size > os.path.getsize(TESTFN)) + with zipfile2.ZipFile(TESTFN,mode="r") as zipf: + self.assertEqual(zipf.comment, b"shorter comment") + + def test_unicode_comment(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with self.assertRaises(TypeError): + zipf.comment = "this is an error" + + def test_change_comment_in_empty_archive(self): + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertFalse(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_change_comment_in_nonempty_archive(self): + with zipfile2.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf: + zipf.writestr("foo.txt", "O, for a Muse of Fire!") + with zipfile2.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf: + self.assertTrue(zipf.filelist) + zipf.comment = b"this is a comment" + with zipfile2.ZipFile(TESTFN, "r") as zipf: + self.assertEqual(zipf.comment, b"this is a comment") + + def test_empty_zipfile(self): + # Check that creating a file in 'w' or 'a' mode and closing without + # adding any files to the archives creates a valid empty ZIP file + zipf = zipfile2.ZipFile(TESTFN, mode="w") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except zipfile.BadZipFile: + self.fail("Unable to create empty ZIP file in 'w' mode") + + zipf = zipfile2.ZipFile(TESTFN, mode="a") + zipf.close() + try: + zipf = zipfile2.ZipFile(TESTFN, mode="r") + except: + self.fail("Unable to create empty ZIP file in 'a' mode") + + def test_open_empty_file(self): + # Issue 1710703: Check that opening a file with less than 22 bytes + # raises a BadZipFile exception (rather than the previously unhelpful + # OSError) + f = open(TESTFN, 'w') + f.close() + self.assertRaises(zipfile.BadZipFile, zipfile2.ZipFile, TESTFN, 'r') + + def test_create_zipinfo_before_1980(self): + self.assertRaises(ValueError, + zipfile.ZipInfo, 'seventies', (1979, 1, 1, 0, 0, 0)) + + def test_create_empty_zipinfo_repr(self): + """Before bpo-26185, repr() on empty ZipInfo object was failing.""" + zi = zipfile.ZipInfo(filename="empty") + self.assertEqual(repr(zi), "") + + def test_create_empty_zipinfo_default_attributes(self): + """Ensure all required attributes are set.""" + zi = zipfile.ZipInfo() + self.assertEqual(zi.orig_filename, "NoName") + self.assertEqual(zi.filename, "NoName") + self.assertEqual(zi.date_time, (1980, 1, 1, 0, 0, 0)) + self.assertEqual(zi.compress_type, zipfile.ZIP_STORED) + self.assertEqual(zi.comment, b"") + self.assertEqual(zi.extra, b"") + self.assertIn(zi.create_system, (0, 3)) + self.assertEqual(zi.create_version, zipfile.DEFAULT_VERSION) + self.assertEqual(zi.extract_version, zipfile.DEFAULT_VERSION) + self.assertEqual(zi.reserved, 0) + self.assertEqual(zi.flag_bits, 0) + self.assertEqual(zi.volume, 0) + self.assertEqual(zi.internal_attr, 0) + self.assertEqual(zi.external_attr, 0) + + # Before bpo-26185, both were missing + self.assertEqual(zi.file_size, 0) + self.assertEqual(zi.compress_size, 0) + + def test_zipfile_with_short_extra_field(self): + """If an extra field in the header is less than 4 bytes, skip it.""" + zipdata = ( + b'PK\x03\x04\x14\x00\x00\x00\x00\x00\x93\x9b\xad@\x8b\x9e' + b'\xd9\xd3\x01\x00\x00\x00\x01\x00\x00\x00\x03\x00\x03\x00ab' + b'c\x00\x00\x00APK\x01\x02\x14\x03\x14\x00\x00\x00\x00' + b'\x00\x93\x9b\xad@\x8b\x9e\xd9\xd3\x01\x00\x00\x00\x01\x00\x00' + b'\x00\x03\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x00\xa4\x81\x00' + b'\x00\x00\x00abc\x00\x00PK\x05\x06\x00\x00\x00\x00' + b'\x01\x00\x01\x003\x00\x00\x00%\x00\x00\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(zipdata), 'r') as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertIsNone(zipf.testzip()) + + def test_open_conflicting_handles(self): + # It's only possible to open one writable file handle at a time + msg1 = b"It's fun to charter an accountant!" + msg2 = b"And sail the wide accountant sea" + msg3 = b"To find, explore the funds offshore" + with zipfile2.ZipFile(TESTFN2, 'w', zipfile.ZIP_STORED) as zipf: + with zipf.open('foo', mode='w') as w2: + w2.write(msg1) + with zipf.open('bar', mode='w') as w1: + with self.assertRaises(ValueError): + zipf.open('handle', mode='w') + with self.assertRaises(ValueError): + zipf.open('foo', mode='r') + with self.assertRaises(ValueError): + zipf.writestr('str', 'abcde') + with self.assertRaises(ValueError): + zipf.write(__file__, 'file') + with self.assertRaises(ValueError): + zipf.close() + w1.write(msg2) + with zipf.open('baz', mode='w') as w2: + w2.write(msg3) + + with zipfile2.ZipFile(TESTFN2, 'r') as zipf: + self.assertEqual(zipf.read('foo'), msg1) + self.assertEqual(zipf.read('bar'), msg2) + self.assertEqual(zipf.read('baz'), msg3) + self.assertEqual(zipf.namelist(), ['foo', 'bar', 'baz']) + + def test_seek_tell(self): + # Test seek functionality + txt = b"Where's Bruce?" + bloc = txt.find(b"Bruce") + # Check seek on a file + with zipfile2.ZipFile(TESTFN, "w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(TESTFN, "r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + # Check seek on memory file + data = io.BytesIO() + with zipfile2.ZipFile(data, mode="w") as zipf: + zipf.writestr("foo.txt", txt) + with zipfile2.ZipFile(data, mode="r") as zipf: + with zipf.open("foo.txt", "r") as fp: + fp.seek(bloc, os.SEEK_SET) + self.assertEqual(fp.tell(), bloc) + fp.seek(-bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), 0) + fp.seek(bloc, os.SEEK_CUR) + self.assertEqual(fp.tell(), bloc) + self.assertEqual(fp.read(5), txt[bloc:bloc+5]) + fp.seek(0, os.SEEK_END) + self.assertEqual(fp.tell(), len(txt)) + fp.seek(0, os.SEEK_SET) + self.assertEqual(fp.tell(), 0) + + @requires_bz2() + def test_decompress_without_3rd_party_library(self): + data = b'PK\x05\x06\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + zip_file = io.BytesIO(data) + with zipfile2.ZipFile(zip_file, 'w', compression=zipfile.ZIP_BZIP2) as zf: + zf.writestr('a.txt', b'a') + with mock.patch('zipfile.bz2', None): + with zipfile2.ZipFile(zip_file) as zf: + self.assertRaises(RuntimeError, zf.extract, 'a.txt') + + @requires_zlib() + def test_full_overlap(self): + data = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e' + b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00a\xed' + b'\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\d\x0b`P' + b'K\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2' + b'\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00aPK' + b'\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e' + b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00bPK\x05' + b'\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00\x00/\x00\x00' + b'\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertEqual(zipf.namelist(), ['a', 'b']) + zi = zipf.getinfo('a') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + zi = zipf.getinfo('b') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + self.assertEqual(len(zipf.read('a')), 1033) + with self.assertRaisesRegex(zipfile.BadZipFile, 'File name.*differ'): + zipf.read('b') + + @requires_zlib() + @unittest.skipIf(sys.version_info < (3, 9, 19), "Behaviour not supported in Python < 3.9.19") + def test_quoted_overlap(self): + data = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05Y\xfc' + b'8\x044\x00\x00\x00(\x04\x00\x00\x01\x00\x00\x00a\x00' + b'\x1f\x00\xe0\xffPK\x03\x04\x14\x00\x00\x00\x08\x00\xa0l' + b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00' + b'\x00\x00b\xed\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\' + b'd\x0b`PK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0' + b'lH\x05Y\xfc8\x044\x00\x00\x00(\x04\x00\x00\x01' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x00aPK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0l' + b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00$\x00\x00\x00' + b'bPK\x05\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00' + b'\x00S\x00\x00\x00\x00\x00' + ) + with zipfile2.ZipFile(io.BytesIO(data), 'r') as zipf: + self.assertEqual(zipf.namelist(), ['a', 'b']) + zi = zipf.getinfo('a') + self.assertEqual(zi.header_offset, 0) + self.assertEqual(zi.compress_size, 52) + self.assertEqual(zi.file_size, 1064) + zi = zipf.getinfo('b') + self.assertEqual(zi.header_offset, 36) + self.assertEqual(zi.compress_size, 16) + self.assertEqual(zi.file_size, 1033) + with self.assertRaisesRegex(zipfile.BadZipFile, 'Overlapped entries'): + zipf.read('a') + self.assertEqual(len(zipf.read('b')), 1033) + + def tearDown(self): + unlink(TESTFN) + unlink(TESTFN2) + + +class AbstractBadCrcTests: + def test_testzip_with_bad_crc(self): + """Tests that files with bad CRCs return their name from testzip.""" + zipdata = self.zip_with_bad_crc + + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + # testzip returns the name of the first corrupt file, or None + self.assertEqual('afile', zipf.testzip()) + + def test_read_with_bad_crc(self): + """Tests that files with bad CRCs raise a BadZipFile exception when read.""" + zipdata = self.zip_with_bad_crc + + # Using ZipFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + self.assertRaises(zipfile.BadZipFile, zipf.read, 'afile') + + # Using ZipExtFile.read() + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + self.assertRaises(zipfile.BadZipFile, corrupt_file.read) + + # Same with small reads (in order to exercise the buffering logic) + with zipfile2.ZipFile(io.BytesIO(zipdata), mode="r") as zipf: + with zipf.open('afile', 'r') as corrupt_file: + corrupt_file.MIN_READ_SIZE = 2 + with self.assertRaises(zipfile.BadZipFile): + while corrupt_file.read(2): + pass + + +class StoredBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_STORED + zip_with_bad_crc = ( + b'PK\003\004\024\0\0\0\0\0 \213\212;:r' + b'\253\377\f\0\0\0\f\0\0\0\005\0\0\000af' + b'ilehello,AworldP' + b'K\001\002\024\003\024\0\0\0\0\0 \213\212;:' + b'r\253\377\f\0\0\0\f\0\0\0\005\0\0\0\0' + b'\0\0\0\0\0\0\0\200\001\0\0\0\000afi' + b'lePK\005\006\0\0\0\0\001\0\001\0003\000' + b'\0\0/\0\0\0\0\0') + +@requires_zlib() +class DeflateBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_DEFLATED + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x00\x00\x00\x08\x00n}\x0c=FA' + b'KE\x10\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\xcbH\xcd\xc9\xc9W(\xcf/\xcaI\xc9\xa0' + b'=\x13\x00PK\x01\x02\x14\x03\x14\x00\x00\x00\x08\x00n' + b'}\x0c=FAKE\x10\x00\x00\x00n\x00\x00\x00\x05' + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00' + b'\x00afilePK\x05\x06\x00\x00\x00\x00\x01\x00' + b'\x01\x003\x00\x00\x003\x00\x00\x00\x00\x00') + +@requires_bz2() +class Bzip2BadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_BZIP2 + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0c\x00nu\x0c=FA' + b'KE8\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ileBZh91AY&SY\xd4\xa8\xca' + b'\x7f\x00\x00\x0f\x11\x80@\x00\x06D\x90\x80 \x00 \xa5' + b'P\xd9!\x03\x03\x13\x13\x13\x89\xa9\xa9\xc2u5:\x9f' + b'\x8b\xb9"\x9c(HjTe?\x80PK\x01\x02\x14' + b'\x03\x14\x03\x00\x00\x0c\x00nu\x0c=FAKE8' + b'\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00\x00\x00\x00' + b'\x00 \x80\x80\x81\x00\x00\x00\x00afilePK' + b'\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00\x00[\x00' + b'\x00\x00\x00\x00') + +@requires_lzma() +class LzmaBadCrcTests(AbstractBadCrcTests, unittest.TestCase): + compression = zipfile.ZIP_LZMA + zip_with_bad_crc = ( + b'PK\x03\x04\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af' + b'ile\t\x04\x05\x00]\x00\x00\x00\x04\x004\x19I' + b'\xee\x8d\xe9\x17\x89:3`\tq!.8\x00PK' + b'\x01\x02\x14\x03\x14\x03\x00\x00\x0e\x00nu\x0c=FA' + b'KE\x1b\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00\x00\x00' + b'\x00\x00\x00\x00 \x80\x80\x81\x00\x00\x00\x00afil' + b'ePK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x003\x00\x00' + b'\x00>\x00\x00\x00\x00\x00') + + +class DecryptionTests(unittest.TestCase): + """Check that ZIP decryption works. Since the library does not + support encryption at the moment, we use a pre-generated encrypted + ZIP file.""" + + data = ( + b'PK\x03\x04\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00\x1a\x00' + b'\x00\x00\x08\x00\x00\x00test.txt\xfa\x10\xa0gly|\xfa-\xc5\xc0=\xf9y' + b'\x18\xe0\xa8r\xb3Z}Lg\xbc\xae\xf9|\x9b\x19\xe4\x8b\xba\xbb)\x8c\xb0\xdbl' + b'PK\x01\x02\x14\x00\x14\x00\x01\x00\x00\x00n\x92i.#y\xef?&\x00\x00\x00' + b'\x1a\x00\x00\x00\x08\x00\x00\x00\x00\x00\x00\x00\x01\x00 \x00\xb6\x81' + b'\x00\x00\x00\x00test.txtPK\x05\x06\x00\x00\x00\x00\x01\x00\x01\x006\x00' + b'\x00\x00L\x00\x00\x00\x00\x00' ) + data2 = ( + b'PK\x03\x04\x14\x00\t\x00\x08\x00\xcf}38xu\xaa\xb2\x14\x00\x00\x00\x00\x02' + b'\x00\x00\x04\x00\x15\x00zeroUT\t\x00\x03\xd6\x8b\x92G\xda\x8b\x92GUx\x04' + b'\x00\xe8\x03\xe8\x03\xc7= 3.9.20") + def test_malformed_paths(self): + """ + Path should handle malformed paths gracefully. + + Paths with leading slashes are not visible. + + Paths with dots are treated like regular files. + """ + data = io.BytesIO() + zf = zipfile2.ZipFile(data, "w") + zf.writestr("/one-slash.txt", b"content") + zf.writestr("//two-slash.txt", b"content") + zf.writestr("../parent.txt", b"content") + zf.filename = '' + root = zipfile.Path(zf) + assert list(map(str, root.iterdir())) == ['../'] + assert root.joinpath('..').joinpath('parent.txt').read_bytes() == b'content' + + def test_unsupported_names(self): + """ + Path segments with special characters are readable. + + On some platforms or file systems, characters like + ``:`` and ``?`` are not allowed, but they are valid + in the zip file. + """ + data = io.BytesIO() + zf = zipfile2.ZipFile(data, "w") + zf.writestr("path?", b"content") + zf.writestr("V: NMS.flac", b"fLaC...") + zf.filename = '' + root = zipfile.Path(zf) + contents = root.iterdir() + assert next(contents).name == 'path?' + assert next(contents).name == 'V: NMS.flac' + assert root.joinpath('V: NMS.flac').read_bytes() == b"fLaC..." + + def test_backslash_not_separator(self): + """ + In a zip file, backslashes are not separators. + """ + data = io.BytesIO() + zf = zipfile2.ZipFile(data, "w") + zf.writestr(DirtyZipInfo.for_name("foo\\bar", zf), b"content") + zf.filename = '' + root = zipfile.Path(zf) + (first,) = root.iterdir() + assert not first.is_dir() + assert first.name == 'foo\\bar' + + +class DirtyZipInfo(zipfile.ZipInfo): + """ + Bypass name sanitization. + """ + + def __init__(self, filename, *args, **kwargs): + super().__init__(filename, *args, **kwargs) + self.filename = filename + + @classmethod + def for_name(cls, name, archive): + """ + Construct the same way that ZipFile.writestr does. + + TODO: extract this functionality and re-use + """ + self = cls(filename=name, date_time=time.localtime(time.time())[:6]) + self.compress_type = archive.compression + self.compress_level = archive.compresslevel + if self.filename.endswith('/'): # pragma: no cover + self.external_attr = 0o40775 << 16 # drwxrwxr-x + self.external_attr |= 0x10 # MS-DOS directory flag + else: + self.external_attr = 0o600 << 16 # ?rw------- + return self + + +if __name__ == "__main__": + unittest.main() diff --git a/cpython-tests/3.9/test_zipfile64.py b/cpython-tests/3.9/test_zipfile64.py new file mode 100644 index 0000000..31d0c92 --- /dev/null +++ b/cpython-tests/3.9/test_zipfile64.py @@ -0,0 +1,148 @@ +# Tests of the full ZIP64 functionality of zipfile +# The support.requires call is the only reason for keeping this separate +# from test_zipfile +from test import support + +# XXX(nnorwitz): disable this test by looking for extralargefile resource, +# which doesn't exist. This test takes over 30 minutes to run in general +# and requires more disk space than most of the buildbots. +support.requires( + 'extralargefile', + 'test requires loads of disk-space bytes and a long time to run' + ) + +import zipfile, os, unittest +import time +import sys + +from tempfile import TemporaryFile + +from test.support import TESTFN, requires_zlib + +import zipfile2 + + +TESTFN2 = TESTFN + "2" + +# How much time in seconds can pass before we print a 'Still working' message. +_PRINT_WORKING_MSG_INTERVAL = 60 + +class TestsWithSourceFile(unittest.TestCase): + def setUp(self): + # Create test data. + line_gen = ("Test of zipfile line %d." % i for i in range(1000000)) + self.data = '\n'.join(line_gen).encode('ascii') + + # And write it to a file. + with open(TESTFN, "wb") as fp: + fp.write(self.data) + + def zipTest(self, f, compression): + # Create the ZIP archive. + with zipfile2.ZipFile(f, "w", compression) as zipfp: + + # It will contain enough copies of self.data to reach about 6 GiB of + # raw data to store. + filecount = 6*1024**3 // len(self.data) + + next_time = time.monotonic() + _PRINT_WORKING_MSG_INTERVAL + for num in range(filecount): + zipfp.writestr("testfn%d" % num, self.data) + # Print still working message since this test can be really slow + if next_time <= time.monotonic(): + next_time = time.monotonic() + _PRINT_WORKING_MSG_INTERVAL + print(( + ' zipTest still writing %d of %d, be patient...' % + (num, filecount)), file=sys.__stdout__) + sys.__stdout__.flush() + + # Read the ZIP archive + with zipfile2.ZipFile(f, "r", compression) as zipfp: + for num in range(filecount): + self.assertEqual(zipfp.read("testfn%d" % num), self.data) + # Print still working message since this test can be really slow + if next_time <= time.monotonic(): + next_time = time.monotonic() + _PRINT_WORKING_MSG_INTERVAL + print(( + ' zipTest still reading %d of %d, be patient...' % + (num, filecount)), file=sys.__stdout__) + sys.__stdout__.flush() + + def testStored(self): + # Try the temp file first. If we do TESTFN2 first, then it hogs + # gigabytes of disk space for the duration of the test. + with TemporaryFile() as f: + self.zipTest(f, zipfile.ZIP_STORED) + self.assertFalse(f.closed) + self.zipTest(TESTFN2, zipfile.ZIP_STORED) + + @requires_zlib() + def testDeflated(self): + # Try the temp file first. If we do TESTFN2 first, then it hogs + # gigabytes of disk space for the duration of the test. + with TemporaryFile() as f: + self.zipTest(f, zipfile.ZIP_DEFLATED) + self.assertFalse(f.closed) + self.zipTest(TESTFN2, zipfile.ZIP_DEFLATED) + + def tearDown(self): + for fname in TESTFN, TESTFN2: + if os.path.exists(fname): + os.remove(fname) + + +class OtherTests(unittest.TestCase): + def testMoreThan64kFiles(self): + # This test checks that more than 64k files can be added to an archive, + # and that the resulting archive can be read properly by ZipFile + with zipfile2.ZipFile(TESTFN, mode="w", allowZip64=True) as zipf: + zipf.debug = 100 + numfiles = (1 << 16) * 3//2 + for i in range(numfiles): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles) + + with zipfile2.ZipFile(TESTFN, mode="r") as zipf2: + self.assertEqual(len(zipf2.namelist()), numfiles) + for i in range(numfiles): + content = zipf2.read("foo%08d" % i).decode('ascii') + self.assertEqual(content, "%d" % (i**3 % 57)) + + def testMoreThan64kFilesAppend(self): + with zipfile2.ZipFile(TESTFN, mode="w", allowZip64=False) as zipf: + zipf.debug = 100 + numfiles = (1 << 16) - 1 + for i in range(numfiles): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles) + with self.assertRaises(zipfile.LargeZipFile): + zipf.writestr("foo%08d" % numfiles, b'') + self.assertEqual(len(zipf.namelist()), numfiles) + + with zipfile2.ZipFile(TESTFN, mode="a", allowZip64=False) as zipf: + zipf.debug = 100 + self.assertEqual(len(zipf.namelist()), numfiles) + with self.assertRaises(zipfile.LargeZipFile): + zipf.writestr("foo%08d" % numfiles, b'') + self.assertEqual(len(zipf.namelist()), numfiles) + + with zipfile2.ZipFile(TESTFN, mode="a", allowZip64=True) as zipf: + zipf.debug = 100 + self.assertEqual(len(zipf.namelist()), numfiles) + numfiles2 = (1 << 16) * 3//2 + for i in range(numfiles, numfiles2): + zipf.writestr("foo%08d" % i, "%d" % (i**3 % 57)) + self.assertEqual(len(zipf.namelist()), numfiles2) + + with zipfile2.ZipFile(TESTFN, mode="r") as zipf2: + self.assertEqual(len(zipf2.namelist()), numfiles2) + for i in range(numfiles2): + content = zipf2.read("foo%08d" % i).decode('ascii') + self.assertEqual(content, "%d" % (i**3 % 57)) + + def tearDown(self): + support.unlink(TESTFN) + support.unlink(TESTFN2) + +if __name__ == "__main__": + unittest.main() diff --git a/cpython-tests/3.9/test_zipimport.py b/cpython-tests/3.9/test_zipimport.py new file mode 100644 index 0000000..5b5f2ae --- /dev/null +++ b/cpython-tests/3.9/test_zipimport.py @@ -0,0 +1,777 @@ +import sys +import os +import marshal +import importlib +import importlib.util +import struct +import time +import unittest +import unittest.mock + +from test import support + +from zipfile import ZipInfo, ZIP_STORED, ZIP_DEFLATED + +import zipimport +import linecache +import doctest +import inspect +import io +from traceback import extract_tb, extract_stack, print_tb +try: + import zlib +except ImportError: + zlib = None + +from zipfile2 import ZipFile + + +test_src = """\ +def get_name(): + return __name__ +def get_file(): + return __file__ +""" +test_co = compile(test_src, "", "exec") +raise_src = 'def do_raise(): raise TypeError\n' + +def make_pyc(co, mtime, size): + data = marshal.dumps(co) + pyc = (importlib.util.MAGIC_NUMBER + + struct.pack("", "exec"), NOW, len(src)) + files = {TESTMOD + pyc_ext: (NOW, pyc), + "some.data": (NOW, "some data")} + self.doTest(pyc_ext, files, TESTMOD) + + def testDefaultOptimizationLevel(self): + # zipimport should use the default optimization level (#28131) + src = """if 1: # indent hack + def test(val): + assert(val) + return val\n""" + files = {TESTMOD + '.py': (NOW, src)} + self.makeZip(files) + sys.path.insert(0, TEMP_ZIP) + mod = importlib.import_module(TESTMOD) + self.assertEqual(mod.test(1), 1) + self.assertRaises(AssertionError, mod.test, False) + + def testImport_WithStuff(self): + # try importing from a zipfile which contains additional + # stuff at the beginning of the file + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, + stuff=b"Some Stuff"*31) + + def assertModuleSource(self, module): + self.assertEqual(inspect.getsource(module), test_src) + + def testGetSource(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, call=self.assertModuleSource) + + def testGetCompiledSource(self): + pyc = make_pyc(compile(test_src, "", "exec"), NOW, len(test_src)) + files = {TESTMOD + ".py": (NOW, test_src), + TESTMOD + pyc_ext: (NOW, pyc)} + self.doTest(pyc_ext, files, TESTMOD, call=self.assertModuleSource) + + def runDoctest(self, callback): + files = {TESTMOD + ".py": (NOW, test_src), + "xyz.txt": (NOW, ">>> log.append(True)\n")} + self.doTest(".py", files, TESTMOD, call=callback) + + def doDoctestFile(self, module): + log = [] + old_master, doctest.master = doctest.master, None + try: + doctest.testfile( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ) + finally: + doctest.master = old_master + self.assertEqual(log,[True]) + + def testDoctestFile(self): + self.runDoctest(self.doDoctestFile) + + def doDoctestSuite(self, module): + log = [] + doctest.DocFileTest( + 'xyz.txt', package=module, module_relative=True, + globs=locals() + ).run() + self.assertEqual(log,[True]) + + def testDoctestSuite(self): + self.runDoctest(self.doDoctestSuite) + + def doTraceback(self, module): + try: + module.do_raise() + except Exception as e: + tb = e.__traceback__.tb_next + + f,lno,n,line = extract_tb(tb, 1)[0] + self.assertEqual(line, raise_src.strip()) + + f,lno,n,line = extract_stack(tb.tb_frame, 1)[0] + self.assertEqual(line, raise_src.strip()) + + s = io.StringIO() + print_tb(tb, 1, s) + self.assertTrue(s.getvalue().endswith(raise_src)) + else: + raise AssertionError("This ought to be impossible") + + def testTraceback(self): + files = {TESTMOD + ".py": (NOW, raise_src)} + self.doTest(None, files, TESTMOD, call=self.doTraceback) + + @unittest.skipIf(support.TESTFN_UNENCODABLE is None, + "need an unencodable filename") + def testUnencodable(self): + filename = support.TESTFN_UNENCODABLE + ".zip" + self.addCleanup(support.unlink, filename) + with ZipFile(filename, "w") as z: + zinfo = ZipInfo(TESTMOD + ".py", time.localtime(NOW)) + zinfo.compress_type = self.compression + z.writestr(zinfo, test_src) + zipimport.zipimporter(filename).load_module(TESTMOD) + + def testBytesPath(self): + filename = support.TESTFN + ".zip" + self.addCleanup(support.unlink, filename) + with ZipFile(filename, "w") as z: + zinfo = ZipInfo(TESTMOD + ".py", time.localtime(NOW)) + zinfo.compress_type = self.compression + z.writestr(zinfo, test_src) + + zipimport.zipimporter(filename) + zipimport.zipimporter(os.fsencode(filename)) + with self.assertRaises(TypeError): + zipimport.zipimporter(bytearray(os.fsencode(filename))) + with self.assertRaises(TypeError): + zipimport.zipimporter(memoryview(os.fsencode(filename))) + + def testComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, comment=b"comment") + + def testBeginningCruftAndComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, stuff=b"cruft" * 64, comment=b"hi") + + def testLargestPossibleComment(self): + files = {TESTMOD + ".py": (NOW, test_src)} + self.doTest(".py", files, TESTMOD, comment=b"c" * ((1 << 16) - 1)) + + +@support.requires_zlib() +class CompressedZipImportTestCase(UncompressedZipImportTestCase): + compression = ZIP_DEFLATED + + +class BadFileZipImportTestCase(unittest.TestCase): + def assertZipFailure(self, filename): + self.assertRaises(zipimport.ZipImportError, + zipimport.zipimporter, filename) + + def testNoFile(self): + self.assertZipFailure('AdfjdkFJKDFJjdklfjs') + + def testEmptyFilename(self): + self.assertZipFailure('') + + def testBadArgs(self): + self.assertRaises(TypeError, zipimport.zipimporter, None) + self.assertRaises(TypeError, zipimport.zipimporter, TESTMOD, kwd=None) + self.assertRaises(TypeError, zipimport.zipimporter, + list(os.fsencode(TESTMOD))) + + def testFilenameTooLong(self): + self.assertZipFailure('A' * 33000) + + def testEmptyFile(self): + support.unlink(TESTMOD) + support.create_empty_file(TESTMOD) + self.assertZipFailure(TESTMOD) + + def testFileUnreadable(self): + support.unlink(TESTMOD) + fd = os.open(TESTMOD, os.O_CREAT, 000) + try: + os.close(fd) + + with self.assertRaises(zipimport.ZipImportError) as cm: + zipimport.zipimporter(TESTMOD) + finally: + # If we leave "the read-only bit" set on Windows, nothing can + # delete TESTMOD, and later tests suffer bogus failures. + os.chmod(TESTMOD, 0o666) + support.unlink(TESTMOD) + + def testNotZipFile(self): + support.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write('a' * 22) + fp.close() + self.assertZipFailure(TESTMOD) + + # XXX: disabled until this works on Big-endian machines + def _testBogusZipFile(self): + support.unlink(TESTMOD) + fp = open(TESTMOD, 'w+') + fp.write(struct.pack('=I', 0x06054B50)) + fp.write('a' * 18) + fp.close() + z = zipimport.zipimporter(TESTMOD) + + try: + self.assertRaises(TypeError, z.find_module, None) + self.assertRaises(TypeError, z.load_module, None) + self.assertRaises(TypeError, z.is_package, None) + self.assertRaises(TypeError, z.get_code, None) + self.assertRaises(TypeError, z.get_data, None) + self.assertRaises(TypeError, z.get_source, None) + + error = zipimport.ZipImportError + self.assertEqual(z.find_module('abc'), None) + + self.assertRaises(error, z.load_module, 'abc') + self.assertRaises(error, z.get_code, 'abc') + self.assertRaises(OSError, z.get_data, 'abc') + self.assertRaises(error, z.get_source, 'abc') + self.assertRaises(error, z.is_package, 'abc') + finally: + zipimport._zip_directory_cache.clear() + + +def tearDownModule(): + support.unlink(TESTMOD) + + +if __name__ == "__main__": + unittest.main() diff --git a/cpython-tests/3.9/zip_cp437_header.zip b/cpython-tests/3.9/zip_cp437_header.zip new file mode 100644 index 0000000..f7c6cf1 Binary files /dev/null and b/cpython-tests/3.9/zip_cp437_header.zip differ diff --git a/cpython-tests/3.9/zipdir.zip b/cpython-tests/3.9/zipdir.zip new file mode 100644 index 0000000..ac21d7a Binary files /dev/null and b/cpython-tests/3.9/zipdir.zip differ diff --git a/cpython-tests/3.9/ziptestdata/README.md b/cpython-tests/3.9/ziptestdata/README.md new file mode 100644 index 0000000..6b9147d --- /dev/null +++ b/cpython-tests/3.9/ziptestdata/README.md @@ -0,0 +1,35 @@ +# Test data for `test_zipfile` + +The test executables in this directory are created manually from header.sh and +the `testdata_module_inside_zip.py` file. You must have infozip's zip utility +installed (`apt install zip` on Debian). + +## Purpose + +These are used to test executable files with an appended zipfile, in a scenario +where the executable is _not_ a Python interpreter itself so our automatic +zipimport machinery (that'd look for `__main__.py`) is not being used. + +## Updating the test executables + +If you update header.sh or the testdata_module_inside_zip.py file, rerun the +commands below. These are expected to be rarely changed, if ever. + +### Standard old format (2.0) zip file + +``` +zip -0 zip2.zip testdata_module_inside_zip.py +cat header.sh zip2.zip >exe_with_zip +rm zip2.zip +``` + +### Modern format (4.5) zip64 file + +Redirecting from stdin forces infozip's zip tool to create a zip64. + +``` +zip -0 zip64.zip +cat header.sh zip64.zip >exe_with_z64 +rm zip64.zip +``` + diff --git a/cpython-tests/3.9/ziptestdata/exe_with_z64 b/cpython-tests/3.9/ziptestdata/exe_with_z64 new file mode 100644 index 0000000..82b03cf Binary files /dev/null and b/cpython-tests/3.9/ziptestdata/exe_with_z64 differ diff --git a/cpython-tests/3.9/ziptestdata/exe_with_zip b/cpython-tests/3.9/ziptestdata/exe_with_zip new file mode 100644 index 0000000..c833cdf Binary files /dev/null and b/cpython-tests/3.9/ziptestdata/exe_with_zip differ diff --git a/cpython-tests/3.9/ziptestdata/header.sh b/cpython-tests/3.9/ziptestdata/header.sh new file mode 100644 index 0000000..52dc91a --- /dev/null +++ b/cpython-tests/3.9/ziptestdata/header.sh @@ -0,0 +1,24 @@ +#!/bin/bash +INTERPRETER_UNDER_TEST="$1" +if [[ ! -x "${INTERPRETER_UNDER_TEST}" ]]; then + echo "Interpreter must be the command line argument." + exit 4 +fi +EXECUTABLE="$0" exec "${INTERPRETER_UNDER_TEST}" -E - <= (3, 12) +PY311 = PYTHON_VERSION >= (3, 11) class TooManyFiles(zipfile.BadZipfile): pass + + +metadata_encoding_warning = 'Python < 3.11 does not support the metadata_encoding' # noqa