"""Test script for the gzip module. """ import array import functools import io import os import pathlib import struct import sys import unittest from subprocess import PIPE, Popen from test import support from test.support import _4G, bigmemtest from test.support.script_helper import assert_python_ok gzip = support.import_module('gzip') data1 = b""" int length=DEFAULTALLOC, err = Z_OK; PyObject *RetVal; int flushmode = Z_FINISH; unsigned long start_total_out; """ data2 = b"""/* zlibmodule.c -- gzip-compatible data compression */ /* See http://www.gzip.org/zlib/ /* See http://www.winimage.com/zLibDll for Windows */ """ TEMPDIR = os.path.abspath(support.TESTFN) + '-gzdir' class UnseekableIO(io.BytesIO): def seekable(self): return False def tell(self): raise io.UnsupportedOperation def seek(self, *args): raise io.UnsupportedOperation class BaseTest(unittest.TestCase): filename = support.TESTFN def setUp(self): support.unlink(self.filename) def tearDown(self): support.unlink(self.filename) class TestGzip(BaseTest): def write_and_read_back(self, data, mode='b'): b_data = bytes(data) with gzip.GzipFile(self.filename, 'w'+mode) as f: l = f.write(data) self.assertEqual(l, len(b_data)) with gzip.GzipFile(self.filename, 'r'+mode) as f: self.assertEqual(f.read(), b_data) def test_write(self): with gzip.GzipFile(self.filename, 'wb') as f: f.write(data1 * 50) # Try flush and fileno. f.flush() f.fileno() if hasattr(os, 'fsync'): os.fsync(f.fileno()) f.close() # Test multiple close() calls. f.close() def test_write_read_with_pathlike_file(self): filename = pathlib.Path(self.filename) with gzip.GzipFile(filename, 'w') as f: f.write(data1 * 50) self.assertIsInstance(f.name, str) with gzip.GzipFile(filename, 'a') as f: f.write(data1) with gzip.GzipFile(filename) as f: d = f.read() self.assertEqual(d, data1 * 51) self.assertIsInstance(f.name, str) # The following test_write_xy methods test that write accepts # the corresponding bytes-like object type as input # and that the data written equals bytes(xy) in all cases. def test_write_memoryview(self): self.write_and_read_back(memoryview(data1 * 50)) m = memoryview(bytes(range(256))) data = m.cast('B', shape=[8,8,4]) self.write_and_read_back(data) def test_write_bytearray(self): self.write_and_read_back(bytearray(data1 * 50)) def test_write_array(self): self.write_and_read_back(array.array('I', data1 * 40)) def test_write_incompatible_type(self): # Test that non-bytes-like types raise TypeError. # Issue #21560: attempts to write incompatible types # should not affect the state of the fileobject with gzip.GzipFile(self.filename, 'wb') as f: with self.assertRaises(TypeError): f.write('') with self.assertRaises(TypeError): f.write([]) f.write(data1) with gzip.GzipFile(self.filename, 'rb') as f: self.assertEqual(f.read(), data1) def test_read(self): self.test_write() # Try reading. with gzip.GzipFile(self.filename, 'r') as f: d = f.read() self.assertEqual(d, data1*50) def test_read1(self): self.test_write() blocks = [] nread = 0 with gzip.GzipFile(self.filename, 'r') as f: while True: d = f.read1() if not d: break blocks.append(d) nread += len(d) # Check that position was updated correctly (see issue10791). self.assertEqual(f.tell(), nread) self.assertEqual(b''.join(blocks), data1 * 50) @bigmemtest(size=_4G, memuse=1) def test_read_large(self, size): # Read chunk size over UINT_MAX should be supported, despite zlib's # limitation per low-level call compressed = gzip.compress(data1, compresslevel=1) f = gzip.GzipFile(fileobj=io.BytesIO(compressed), mode='rb') self.assertEqual(f.read(size), data1) def test_io_on_closed_object(self): # Test that I/O operations on closed GzipFile objects raise a # ValueError, just like the corresponding functions on file objects. # Write to a file, open it for reading, then close it. self.test_write() f = gzip.GzipFile(self.filename, 'r') fileobj = f.fileobj self.assertFalse(fileobj.closed) f.close() self.assertTrue(fileobj.closed) with self.assertRaises(ValueError): f.read(1) with self.assertRaises(ValueError): f.seek(0) with self.assertRaises(ValueError): f.tell() # Open the file for writing, then close it. f = gzip.GzipFile(self.filename, 'w') fileobj = f.fileobj self.assertFalse(fileobj.closed) f.close() self.assertTrue(fileobj.closed) with self.assertRaises(ValueError): f.write(b'') with self.assertRaises(ValueError): f.flush() def test_append(self): self.test_write() # Append to the previous file with gzip.GzipFile(self.filename, 'ab') as f: f.write(data2 * 15) with gzip.GzipFile(self.filename, 'rb') as f: d = f.read() self.assertEqual(d, (data1*50) + (data2*15)) def test_many_append(self): # Bug #1074261 was triggered when reading a file that contained # many, many members. Create such a file and verify that reading it # works. with gzip.GzipFile(self.filename, 'wb', 9) as f: f.write(b'a') for i in range(0, 200): with gzip.GzipFile(self.filename, "ab", 9) as f: # append f.write(b'a') # Try reading the file with gzip.GzipFile(self.filename, "rb") as zgfile: contents = b"" while 1: ztxt = zgfile.read(8192) contents += ztxt if not ztxt: break self.assertEqual(contents, b'a'*201) def test_exclusive_write(self): with gzip.GzipFile(self.filename, 'xb') as f: f.write(data1 * 50) with gzip.GzipFile(self.filename, 'rb') as f: self.assertEqual(f.read(), data1 * 50) with self.assertRaises(FileExistsError): gzip.GzipFile(self.filename, 'xb') def test_buffered_reader(self): # Issue #7471: a GzipFile can be wrapped in a BufferedReader for # performance. self.test_write() with gzip.GzipFile(self.filename, 'rb') as f: with io.BufferedReader(f) as r: lines = [line for line in r] self.assertEqual(lines, 50 * data1.splitlines(keepends=True)) def test_readline(self): self.test_write() # Try .readline() with varying line lengths with gzip.GzipFile(self.filename, 'rb') as f: line_length = 0 while 1: L = f.readline(line_length) if not L and line_length != 0: break self.assertTrue(len(L) <= line_length) line_length = (line_length + 1) % 50 def test_readlines(self): self.test_write() # Try .readlines() with gzip.GzipFile(self.filename, 'rb') as f: L = f.readlines() with gzip.GzipFile(self.filename, 'rb') as f: while 1: L = f.readlines(150) if L == []: break def test_seek_read(self): self.test_write() # Try seek, read test with gzip.GzipFile(self.filename) as f: while 1: oldpos = f.tell() line1 = f.readline() if not line1: break newpos = f.tell() f.seek(oldpos) # negative seek if len(line1)>10: amount = 10 else: amount = len(line1) line2 = f.read(amount) self.assertEqual(line1[:amount], line2) f.seek(newpos) # positive seek def test_seek_whence(self): self.test_write() # Try seek(whence=1), read test with gzip.GzipFile(self.filename) as f: f.read(10) f.seek(10, whence=1) y = f.read(10) self.assertEqual(y, data1[20:30]) def test_seek_write(self): # Try seek, write test with gzip.GzipFile(self.filename, 'w') as f: for pos in range(0, 256, 16): f.seek(pos) f.write(b'GZ\n') def test_mode(self): self.test_write() with gzip.GzipFile(self.filename, 'r') as f: self.assertEqual(f.myfileobj.mode, 'rb') support.unlink(self.filename) with gzip.GzipFile(self.filename, 'x') as f: self.assertEqual(f.myfileobj.mode, 'xb') def test_1647484(self): for mode in ('wb', 'rb'): with gzip.GzipFile(self.filename, mode) as f: self.assertTrue(hasattr(f, "name")) self.assertEqual(f.name, self.filename) def test_paddedfile_getattr(self): self.test_write() with gzip.GzipFile(self.filename, 'rb') as f: self.assertTrue(hasattr(f.fileobj, "name")) self.assertEqual(f.fileobj.name, self.filename) def test_mtime(self): mtime = 123456789 with gzip.GzipFile(self.filename, 'w', mtime = mtime) as fWrite: fWrite.write(data1) with gzip.GzipFile(self.filename) as fRead: self.assertTrue(hasattr(fRead, 'mtime')) self.assertIsNone(fRead.mtime) dataRead = fRead.read() self.assertEqual(dataRead, data1) self.assertEqual(fRead.mtime, mtime) def test_metadata(self): mtime = 123456789 with gzip.GzipFile(self.filename, 'w', mtime = mtime) as fWrite: fWrite.write(data1) with open(self.filename, 'rb') as fRead: # see RFC 1952: http://www.faqs.org/rfcs/rfc1952.html idBytes = fRead.read(2) self.assertEqual(idBytes, b'\x1f\x8b') # gzip ID cmByte = fRead.read(1) self.assertEqual(cmByte, b'\x08') # deflate flagsByte = fRead.read(1) self.assertEqual(flagsByte, b'\x08') # only the FNAME flag is set mtimeBytes = fRead.read(4) self.assertEqual(mtimeBytes, struct.pack('