Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 1 | import unittest |
Benjamin Peterson | ee8712c | 2008-05-20 21:35:26 +0000 | [diff] [blame] | 2 | from test import support |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 3 | import binascii |
Serhiy Storchaka | d7a4415 | 2015-11-12 11:23:04 +0200 | [diff] [blame] | 4 | import pickle |
Andrew M. Kuchling | 9a0f98e | 2001-02-21 02:17:01 +0000 | [diff] [blame] | 5 | import random |
Antoine Pitrou | f3d2275 | 2011-02-21 18:09:00 +0000 | [diff] [blame] | 6 | import sys |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 7 | from test.support import bigmemtest, _1G, _4G |
Andrew M. Kuchling | 9a0f98e | 2001-02-21 02:17:01 +0000 | [diff] [blame] | 8 | |
R. David Murray | a21e4ca | 2009-03-31 23:16:50 +0000 | [diff] [blame] | 9 | zlib = support.import_module('zlib') |
| 10 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 11 | requires_Compress_copy = unittest.skipUnless( |
| 12 | hasattr(zlib.compressobj(), "copy"), |
| 13 | 'requires Compress.copy()') |
| 14 | requires_Decompress_copy = unittest.skipUnless( |
| 15 | hasattr(zlib.decompressobj(), "copy"), |
| 16 | 'requires Decompress.copy()') |
| 17 | |
Andrew M. Kuchling | 9a0f98e | 2001-02-21 02:17:01 +0000 | [diff] [blame] | 18 | |
Nadeem Vawda | 64d25dd | 2011-09-12 00:04:13 +0200 | [diff] [blame] | 19 | class VersionTestCase(unittest.TestCase): |
| 20 | |
| 21 | def test_library_version(self): |
Nadeem Vawda | 131c707 | 2012-01-25 23:16:50 +0200 | [diff] [blame] | 22 | # Test that the major version of the actual library in use matches the |
| 23 | # major version that we were compiled against. We can't guarantee that |
| 24 | # the minor versions will match (even on the machine on which the module |
| 25 | # was compiled), and the API is stable between minor versions, so |
Nadeem Vawda | d770fe4 | 2012-01-28 17:32:47 +0200 | [diff] [blame] | 26 | # testing only the major versions avoids spurious failures. |
Nadeem Vawda | 131c707 | 2012-01-25 23:16:50 +0200 | [diff] [blame] | 27 | self.assertEqual(zlib.ZLIB_RUNTIME_VERSION[0], zlib.ZLIB_VERSION[0]) |
Nadeem Vawda | 64d25dd | 2011-09-12 00:04:13 +0200 | [diff] [blame] | 28 | |
| 29 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 30 | class ChecksumTestCase(unittest.TestCase): |
| 31 | # checksum test cases |
| 32 | def test_crc32start(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 33 | self.assertEqual(zlib.crc32(b""), zlib.crc32(b"", 0)) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 34 | self.assertTrue(zlib.crc32(b"abc", 0xffffffff)) |
Andrew M. Kuchling | fcfc8d5 | 2001-08-10 15:50:11 +0000 | [diff] [blame] | 35 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 36 | def test_crc32empty(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 37 | self.assertEqual(zlib.crc32(b"", 0), 0) |
| 38 | self.assertEqual(zlib.crc32(b"", 1), 1) |
| 39 | self.assertEqual(zlib.crc32(b"", 432), 432) |
Andrew M. Kuchling | 9a0f98e | 2001-02-21 02:17:01 +0000 | [diff] [blame] | 40 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 41 | def test_adler32start(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 42 | self.assertEqual(zlib.adler32(b""), zlib.adler32(b"", 1)) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 43 | self.assertTrue(zlib.adler32(b"abc", 0xffffffff)) |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 44 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 45 | def test_adler32empty(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 46 | self.assertEqual(zlib.adler32(b"", 0), 0) |
| 47 | self.assertEqual(zlib.adler32(b"", 1), 1) |
| 48 | self.assertEqual(zlib.adler32(b"", 432), 432) |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 49 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 50 | def assertEqual32(self, seen, expected): |
| 51 | # 32-bit values masked -- checksums on 32- vs 64- bit machines |
| 52 | # This is important if bit 31 (0x08000000L) is set. |
Guido van Rossum | e2a383d | 2007-01-15 16:59:06 +0000 | [diff] [blame] | 53 | self.assertEqual(seen & 0x0FFFFFFFF, expected & 0x0FFFFFFFF) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 54 | |
| 55 | def test_penguins(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 56 | self.assertEqual32(zlib.crc32(b"penguin", 0), 0x0e5c1a120) |
| 57 | self.assertEqual32(zlib.crc32(b"penguin", 1), 0x43b6aa94) |
| 58 | self.assertEqual32(zlib.adler32(b"penguin", 0), 0x0bcf02f6) |
| 59 | self.assertEqual32(zlib.adler32(b"penguin", 1), 0x0bd602f7) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 60 | |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 61 | self.assertEqual(zlib.crc32(b"penguin"), zlib.crc32(b"penguin", 0)) |
| 62 | self.assertEqual(zlib.adler32(b"penguin"),zlib.adler32(b"penguin",1)) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 63 | |
Gregory P. Smith | ab0d8a1 | 2008-03-17 20:24:09 +0000 | [diff] [blame] | 64 | def test_crc32_adler32_unsigned(self): |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 65 | foo = b'abcdefghijklmnop' |
Gregory P. Smith | ab0d8a1 | 2008-03-17 20:24:09 +0000 | [diff] [blame] | 66 | # explicitly test signed behavior |
Gregory P. Smith | 2727503 | 2008-03-20 06:20:09 +0000 | [diff] [blame] | 67 | self.assertEqual(zlib.crc32(foo), 2486878355) |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 68 | self.assertEqual(zlib.crc32(b'spam'), 1138425661) |
Gregory P. Smith | ab0d8a1 | 2008-03-17 20:24:09 +0000 | [diff] [blame] | 69 | self.assertEqual(zlib.adler32(foo+foo), 3573550353) |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 70 | self.assertEqual(zlib.adler32(b'spam'), 72286642) |
Gregory P. Smith | ab0d8a1 | 2008-03-17 20:24:09 +0000 | [diff] [blame] | 71 | |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 72 | def test_same_as_binascii_crc32(self): |
Martin v. Löwis | 15b16a3 | 2008-12-02 06:00:15 +0000 | [diff] [blame] | 73 | foo = b'abcdefghijklmnop' |
Gregory P. Smith | 2727503 | 2008-03-20 06:20:09 +0000 | [diff] [blame] | 74 | crc = 2486878355 |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 75 | self.assertEqual(binascii.crc32(foo), crc) |
| 76 | self.assertEqual(zlib.crc32(foo), crc) |
Martin v. Löwis | 15b16a3 | 2008-12-02 06:00:15 +0000 | [diff] [blame] | 77 | self.assertEqual(binascii.crc32(b'spam'), zlib.crc32(b'spam')) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 78 | |
| 79 | |
Antoine Pitrou | f3d2275 | 2011-02-21 18:09:00 +0000 | [diff] [blame] | 80 | # Issue #10276 - check that inputs >=4GB are handled correctly. |
| 81 | class ChecksumBigBufferTestCase(unittest.TestCase): |
| 82 | |
Nadeem Vawda | bc8c817 | 2012-02-23 14:16:15 +0200 | [diff] [blame] | 83 | @bigmemtest(size=_4G + 4, memuse=1, dry_run=False) |
| 84 | def test_big_buffer(self, size): |
Nadeem Vawda | b063a48 | 2012-02-23 13:36:25 +0200 | [diff] [blame] | 85 | data = b"nyan" * (_1G + 1) |
| 86 | self.assertEqual(zlib.crc32(data), 1044521549) |
| 87 | self.assertEqual(zlib.adler32(data), 2256789997) |
Antoine Pitrou | f3d2275 | 2011-02-21 18:09:00 +0000 | [diff] [blame] | 88 | |
Christian Heimes | b186d00 | 2008-03-18 15:15:01 +0000 | [diff] [blame] | 89 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 90 | class ExceptionTestCase(unittest.TestCase): |
| 91 | # make sure we generate some expected errors |
Guido van Rossum | 8ce8a78 | 2007-11-01 19:42:39 +0000 | [diff] [blame] | 92 | def test_badlevel(self): |
| 93 | # specifying compression level out of range causes an error |
| 94 | # (but -1 is Z_DEFAULT_COMPRESSION and apparently the zlib |
| 95 | # accepts 0 too) |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 96 | self.assertRaises(zlib.error, zlib.compress, b'ERROR', 10) |
| 97 | |
| 98 | def test_badargs(self): |
| 99 | self.assertRaises(TypeError, zlib.adler32) |
| 100 | self.assertRaises(TypeError, zlib.crc32) |
| 101 | self.assertRaises(TypeError, zlib.compress) |
| 102 | self.assertRaises(TypeError, zlib.decompress) |
| 103 | for arg in (42, None, '', 'abc', (), []): |
| 104 | self.assertRaises(TypeError, zlib.adler32, arg) |
| 105 | self.assertRaises(TypeError, zlib.crc32, arg) |
| 106 | self.assertRaises(TypeError, zlib.compress, arg) |
| 107 | self.assertRaises(TypeError, zlib.decompress, arg) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 108 | |
| 109 | def test_badcompressobj(self): |
| 110 | # verify failure on building compress object with bad params |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 111 | self.assertRaises(ValueError, zlib.compressobj, 1, zlib.DEFLATED, 0) |
Guido van Rossum | 8ce8a78 | 2007-11-01 19:42:39 +0000 | [diff] [blame] | 112 | # specifying total bits too large causes an error |
| 113 | self.assertRaises(ValueError, |
| 114 | zlib.compressobj, 1, zlib.DEFLATED, zlib.MAX_WBITS + 1) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 115 | |
| 116 | def test_baddecompressobj(self): |
| 117 | # verify failure on building decompress object with bad params |
Antoine Pitrou | 90ee4df | 2010-04-06 17:23:13 +0000 | [diff] [blame] | 118 | self.assertRaises(ValueError, zlib.decompressobj, -1) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 119 | |
Christian Heimes | 5e69685 | 2008-04-09 08:37:03 +0000 | [diff] [blame] | 120 | def test_decompressobj_badflush(self): |
| 121 | # verify failure on calling decompressobj.flush with bad params |
| 122 | self.assertRaises(ValueError, zlib.decompressobj().flush, 0) |
| 123 | self.assertRaises(ValueError, zlib.decompressobj().flush, -1) |
| 124 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 125 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 126 | class BaseCompressTestCase(object): |
| 127 | def check_big_compress_buffer(self, size, compress_func): |
| 128 | _1M = 1024 * 1024 |
| 129 | fmt = "%%0%dx" % (2 * _1M) |
| 130 | # Generate 10MB worth of random, and expand it by repeating it. |
| 131 | # The assumption is that zlib's memory is not big enough to exploit |
| 132 | # such spread out redundancy. |
| 133 | data = b''.join([random.getrandbits(8 * _1M).to_bytes(_1M, 'little') |
| 134 | for i in range(10)]) |
| 135 | data = data * (size // len(data) + 1) |
| 136 | try: |
| 137 | compress_func(data) |
| 138 | finally: |
| 139 | # Release memory |
| 140 | data = None |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 141 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 142 | def check_big_decompress_buffer(self, size, decompress_func): |
| 143 | data = b'x' * size |
| 144 | try: |
| 145 | compressed = zlib.compress(data, 1) |
| 146 | finally: |
| 147 | # Release memory |
| 148 | data = None |
| 149 | data = decompress_func(compressed) |
| 150 | # Sanity check |
| 151 | try: |
| 152 | self.assertEqual(len(data), size) |
| 153 | self.assertEqual(len(data.strip(b'x')), 0) |
| 154 | finally: |
| 155 | data = None |
| 156 | |
| 157 | |
| 158 | class CompressTestCase(BaseCompressTestCase, unittest.TestCase): |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 159 | # Test compression in one go (whole message compression) |
| 160 | def test_speech(self): |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 161 | x = zlib.compress(HAMLET_SCENE) |
| 162 | self.assertEqual(zlib.decompress(x), HAMLET_SCENE) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 163 | |
| 164 | def test_speech128(self): |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 165 | # compress more data |
| 166 | data = HAMLET_SCENE * 128 |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 167 | x = zlib.compress(data) |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 168 | self.assertEqual(zlib.compress(bytearray(data)), x) |
| 169 | for ob in x, bytearray(x): |
| 170 | self.assertEqual(zlib.decompress(ob), data) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 171 | |
Antoine Pitrou | 53b2166 | 2010-05-11 23:46:02 +0000 | [diff] [blame] | 172 | def test_incomplete_stream(self): |
| 173 | # An useful error message is given |
| 174 | x = zlib.compress(HAMLET_SCENE) |
Ezio Melotti | ed3a7d2 | 2010-12-01 02:32:32 +0000 | [diff] [blame] | 175 | self.assertRaisesRegex(zlib.error, |
Antoine Pitrou | 53b2166 | 2010-05-11 23:46:02 +0000 | [diff] [blame] | 176 | "Error -5 while decompressing data: incomplete or truncated stream", |
| 177 | zlib.decompress, x[:-1]) |
| 178 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 179 | # Memory use of the following functions takes into account overallocation |
| 180 | |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 181 | @bigmemtest(size=_1G + 1024 * 1024, memuse=3) |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 182 | def test_big_compress_buffer(self, size): |
| 183 | compress = lambda s: zlib.compress(s, 1) |
| 184 | self.check_big_compress_buffer(size, compress) |
| 185 | |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 186 | @bigmemtest(size=_1G + 1024 * 1024, memuse=2) |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 187 | def test_big_decompress_buffer(self, size): |
| 188 | self.check_big_decompress_buffer(size, zlib.decompress) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 189 | |
Nadeem Vawda | 197e22c | 2012-02-23 14:23:17 +0200 | [diff] [blame] | 190 | @bigmemtest(size=_4G + 100, memuse=1, dry_run=False) |
Victor Stinner | 8848c7a | 2011-01-04 02:07:36 +0000 | [diff] [blame] | 191 | def test_length_overflow(self, size): |
Victor Stinner | 8848c7a | 2011-01-04 02:07:36 +0000 | [diff] [blame] | 192 | data = b'x' * size |
| 193 | try: |
| 194 | self.assertRaises(OverflowError, zlib.compress, data, 1) |
Nadeem Vawda | 154bdf9 | 2011-05-14 23:07:36 +0200 | [diff] [blame] | 195 | self.assertRaises(OverflowError, zlib.decompress, data) |
Victor Stinner | 8848c7a | 2011-01-04 02:07:36 +0000 | [diff] [blame] | 196 | finally: |
| 197 | data = None |
| 198 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 199 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 200 | class CompressObjectTestCase(BaseCompressTestCase, unittest.TestCase): |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 201 | # Test compression object |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 202 | def test_pair(self): |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 203 | # straightforward compress/decompress objects |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 204 | datasrc = HAMLET_SCENE * 128 |
| 205 | datazip = zlib.compress(datasrc) |
| 206 | # should compress both bytes and bytearray data |
| 207 | for data in (datasrc, bytearray(datasrc)): |
| 208 | co = zlib.compressobj() |
| 209 | x1 = co.compress(data) |
| 210 | x2 = co.flush() |
| 211 | self.assertRaises(zlib.error, co.flush) # second flush should not work |
| 212 | self.assertEqual(x1 + x2, datazip) |
| 213 | for v1, v2 in ((x1, x2), (bytearray(x1), bytearray(x2))): |
| 214 | dco = zlib.decompressobj() |
| 215 | y1 = dco.decompress(v1 + v2) |
| 216 | y2 = dco.flush() |
| 217 | self.assertEqual(data, y1 + y2) |
| 218 | self.assertIsInstance(dco.unconsumed_tail, bytes) |
| 219 | self.assertIsInstance(dco.unused_data, bytes) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 220 | |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 221 | def test_compressoptions(self): |
| 222 | # specify lots of options to compressobj() |
| 223 | level = 2 |
| 224 | method = zlib.DEFLATED |
| 225 | wbits = -12 |
Martin Panter | bf19d16 | 2015-09-09 01:01:13 +0000 | [diff] [blame] | 226 | memLevel = 9 |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 227 | strategy = zlib.Z_FILTERED |
Martin Panter | bf19d16 | 2015-09-09 01:01:13 +0000 | [diff] [blame] | 228 | co = zlib.compressobj(level, method, wbits, memLevel, strategy) |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 229 | x1 = co.compress(HAMLET_SCENE) |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 230 | x2 = co.flush() |
| 231 | dco = zlib.decompressobj(wbits) |
| 232 | y1 = dco.decompress(x1 + x2) |
| 233 | y2 = dco.flush() |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 234 | self.assertEqual(HAMLET_SCENE, y1 + y2) |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 235 | |
Martin Panter | bf19d16 | 2015-09-09 01:01:13 +0000 | [diff] [blame] | 236 | # keyword arguments should also be supported |
| 237 | zlib.compressobj(level=level, method=method, wbits=wbits, |
| 238 | memLevel=memLevel, strategy=strategy, zdict=b"") |
| 239 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 240 | def test_compressincremental(self): |
| 241 | # compress object in steps, decompress object as one-shot |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 242 | data = HAMLET_SCENE * 128 |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 243 | co = zlib.compressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 244 | bufs = [] |
| 245 | for i in range(0, len(data), 256): |
| 246 | bufs.append(co.compress(data[i:i+256])) |
| 247 | bufs.append(co.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 248 | combuf = b''.join(bufs) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 249 | |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 250 | dco = zlib.decompressobj() |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 251 | y1 = dco.decompress(b''.join(bufs)) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 252 | y2 = dco.flush() |
| 253 | self.assertEqual(data, y1 + y2) |
| 254 | |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 255 | def test_decompinc(self, flush=False, source=None, cx=256, dcx=64): |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 256 | # compress object in steps, decompress object in steps |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 257 | source = source or HAMLET_SCENE |
| 258 | data = source * 128 |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 259 | co = zlib.compressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 260 | bufs = [] |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 261 | for i in range(0, len(data), cx): |
| 262 | bufs.append(co.compress(data[i:i+cx])) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 263 | bufs.append(co.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 264 | combuf = b''.join(bufs) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 265 | |
Gregory P. Smith | 693fc46 | 2008-09-06 20:13:06 +0000 | [diff] [blame] | 266 | decombuf = zlib.decompress(combuf) |
| 267 | # Test type of return value |
Ezio Melotti | e961593 | 2010-01-24 19:26:24 +0000 | [diff] [blame] | 268 | self.assertIsInstance(decombuf, bytes) |
Gregory P. Smith | 693fc46 | 2008-09-06 20:13:06 +0000 | [diff] [blame] | 269 | |
| 270 | self.assertEqual(data, decombuf) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 271 | |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 272 | dco = zlib.decompressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 273 | bufs = [] |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 274 | for i in range(0, len(combuf), dcx): |
| 275 | bufs.append(dco.decompress(combuf[i:i+dcx])) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 276 | self.assertEqual(b'', dco.unconsumed_tail, ######## |
| 277 | "(A) uct should be b'': not %d long" % |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 278 | len(dco.unconsumed_tail)) |
Amaury Forgeot d'Arc | e43d33a | 2008-07-02 20:50:16 +0000 | [diff] [blame] | 279 | self.assertEqual(b'', dco.unused_data) |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 280 | if flush: |
| 281 | bufs.append(dco.flush()) |
| 282 | else: |
| 283 | while True: |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 284 | chunk = dco.decompress(b'') |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 285 | if chunk: |
| 286 | bufs.append(chunk) |
| 287 | else: |
| 288 | break |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 289 | self.assertEqual(b'', dco.unconsumed_tail, ######## |
| 290 | "(B) uct should be b'': not %d long" % |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 291 | len(dco.unconsumed_tail)) |
Amaury Forgeot d'Arc | e43d33a | 2008-07-02 20:50:16 +0000 | [diff] [blame] | 292 | self.assertEqual(b'', dco.unused_data) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 293 | self.assertEqual(data, b''.join(bufs)) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 294 | # Failure means: "decompressobj with init options failed" |
| 295 | |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 296 | def test_decompincflush(self): |
| 297 | self.test_decompinc(flush=True) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 298 | |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 299 | def test_decompimax(self, source=None, cx=256, dcx=64): |
| 300 | # compress in steps, decompress in length-restricted steps |
| 301 | source = source or HAMLET_SCENE |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 302 | # Check a decompression object with max_length specified |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 303 | data = source * 128 |
| 304 | co = zlib.compressobj() |
| 305 | bufs = [] |
| 306 | for i in range(0, len(data), cx): |
| 307 | bufs.append(co.compress(data[i:i+cx])) |
| 308 | bufs.append(co.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 309 | combuf = b''.join(bufs) |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 310 | self.assertEqual(data, zlib.decompress(combuf), |
| 311 | 'compressed data failure') |
| 312 | |
| 313 | dco = zlib.decompressobj() |
| 314 | bufs = [] |
| 315 | cb = combuf |
| 316 | while cb: |
| 317 | #max_length = 1 + len(cb)//10 |
| 318 | chunk = dco.decompress(cb, dcx) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 319 | self.assertFalse(len(chunk) > dcx, |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 320 | 'chunk too big (%d>%d)' % (len(chunk), dcx)) |
| 321 | bufs.append(chunk) |
| 322 | cb = dco.unconsumed_tail |
| 323 | bufs.append(dco.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 324 | self.assertEqual(data, b''.join(bufs), 'Wrong data retrieved') |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 325 | |
| 326 | def test_decompressmaxlen(self, flush=False): |
| 327 | # Check a decompression object with max_length specified |
| 328 | data = HAMLET_SCENE * 128 |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 329 | co = zlib.compressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 330 | bufs = [] |
| 331 | for i in range(0, len(data), 256): |
| 332 | bufs.append(co.compress(data[i:i+256])) |
| 333 | bufs.append(co.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 334 | combuf = b''.join(bufs) |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 335 | self.assertEqual(data, zlib.decompress(combuf), |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 336 | 'compressed data failure') |
| 337 | |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 338 | dco = zlib.decompressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 339 | bufs = [] |
| 340 | cb = combuf |
| 341 | while cb: |
Guido van Rossum | f359410 | 2003-02-27 18:39:18 +0000 | [diff] [blame] | 342 | max_length = 1 + len(cb)//10 |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 343 | chunk = dco.decompress(cb, max_length) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 344 | self.assertFalse(len(chunk) > max_length, |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 345 | 'chunk too big (%d>%d)' % (len(chunk),max_length)) |
| 346 | bufs.append(chunk) |
| 347 | cb = dco.unconsumed_tail |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 348 | if flush: |
| 349 | bufs.append(dco.flush()) |
| 350 | else: |
| 351 | while chunk: |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 352 | chunk = dco.decompress(b'', max_length) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 353 | self.assertFalse(len(chunk) > max_length, |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 354 | 'chunk too big (%d>%d)' % (len(chunk),max_length)) |
| 355 | bufs.append(chunk) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 356 | self.assertEqual(data, b''.join(bufs), 'Wrong data retrieved') |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 357 | |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 358 | def test_decompressmaxlenflush(self): |
| 359 | self.test_decompressmaxlen(flush=True) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 360 | |
| 361 | def test_maxlenmisc(self): |
| 362 | # Misc tests of max_length |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 363 | dco = zlib.decompressobj() |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 364 | self.assertRaises(ValueError, dco.decompress, b"", -1) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 365 | self.assertEqual(b'', dco.unconsumed_tail) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 366 | |
Nadeem Vawda | 7619e88 | 2011-05-14 14:05:20 +0200 | [diff] [blame] | 367 | def test_clear_unconsumed_tail(self): |
| 368 | # Issue #12050: calling decompress() without providing max_length |
| 369 | # should clear the unconsumed_tail attribute. |
| 370 | cdata = b"x\x9cKLJ\x06\x00\x02M\x01" # "abc" |
| 371 | dco = zlib.decompressobj() |
| 372 | ddata = dco.decompress(cdata, 1) |
| 373 | ddata += dco.decompress(dco.unconsumed_tail) |
| 374 | self.assertEqual(dco.unconsumed_tail, b"") |
| 375 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 376 | def test_flushes(self): |
| 377 | # Test flush() with the various options, using all the |
| 378 | # different levels in order to provide more variations. |
| 379 | sync_opt = ['Z_NO_FLUSH', 'Z_SYNC_FLUSH', 'Z_FULL_FLUSH'] |
| 380 | sync_opt = [getattr(zlib, opt) for opt in sync_opt |
| 381 | if hasattr(zlib, opt)] |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 382 | data = HAMLET_SCENE * 8 |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 383 | |
| 384 | for sync in sync_opt: |
| 385 | for level in range(10): |
| 386 | obj = zlib.compressobj( level ) |
| 387 | a = obj.compress( data[:3000] ) |
| 388 | b = obj.flush( sync ) |
| 389 | c = obj.compress( data[3000:] ) |
| 390 | d = obj.flush() |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 391 | self.assertEqual(zlib.decompress(b''.join([a,b,c,d])), |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 392 | data, ("Decompress failed: flush " |
| 393 | "mode=%i, level=%i") % (sync, level)) |
| 394 | del obj |
| 395 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 396 | @unittest.skipUnless(hasattr(zlib, 'Z_SYNC_FLUSH'), |
| 397 | 'requires zlib.Z_SYNC_FLUSH') |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 398 | def test_odd_flush(self): |
| 399 | # Test for odd flushing bugs noted in 2.0, and hopefully fixed in 2.1 |
| 400 | import random |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 401 | # Testing on 17K of "random" data |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 402 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 403 | # Create compressor and decompressor objects |
| 404 | co = zlib.compressobj(zlib.Z_BEST_COMPRESSION) |
| 405 | dco = zlib.decompressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 406 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 407 | # Try 17K of data |
| 408 | # generate random data stream |
| 409 | try: |
| 410 | # In 2.3 and later, WichmannHill is the RNG of the bug report |
| 411 | gen = random.WichmannHill() |
| 412 | except AttributeError: |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 413 | try: |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 414 | # 2.2 called it Random |
| 415 | gen = random.Random() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 416 | except AttributeError: |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 417 | # others might simply have a single RNG |
| 418 | gen = random |
| 419 | gen.seed(1) |
| 420 | data = genblock(1, 17 * 1024, generator=gen) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 421 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 422 | # compress, sync-flush, and decompress |
| 423 | first = co.compress(data) |
| 424 | second = co.flush(zlib.Z_SYNC_FLUSH) |
| 425 | expanded = dco.decompress(first + second) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 426 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 427 | # if decompressed data is different from the input data, choke. |
| 428 | self.assertEqual(expanded, data, "17K random source doesn't match") |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 429 | |
Andrew M. Kuchling | 3b585b3 | 2004-12-28 20:10:48 +0000 | [diff] [blame] | 430 | def test_empty_flush(self): |
| 431 | # Test that calling .flush() on unused objects works. |
| 432 | # (Bug #1083110 -- calling .flush() on decompress objects |
| 433 | # caused a core dump.) |
| 434 | |
| 435 | co = zlib.compressobj(zlib.Z_BEST_COMPRESSION) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 436 | self.assertTrue(co.flush()) # Returns a zlib header |
Andrew M. Kuchling | 3b585b3 | 2004-12-28 20:10:48 +0000 | [diff] [blame] | 437 | dco = zlib.decompressobj() |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 438 | self.assertEqual(dco.flush(), b"") # Returns nothing |
Tim Peters | 5a9fb3c | 2005-01-07 16:01:32 +0000 | [diff] [blame] | 439 | |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 440 | def test_dictionary(self): |
| 441 | h = HAMLET_SCENE |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 442 | # Build a simulated dictionary out of the words in HAMLET. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 443 | words = h.split() |
| 444 | random.shuffle(words) |
| 445 | zdict = b''.join(words) |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 446 | # Use it to compress HAMLET. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 447 | co = zlib.compressobj(zdict=zdict) |
| 448 | cd = co.compress(h) + co.flush() |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 449 | # Verify that it will decompress with the dictionary. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 450 | dco = zlib.decompressobj(zdict=zdict) |
| 451 | self.assertEqual(dco.decompress(cd) + dco.flush(), h) |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 452 | # Verify that it fails when not given the dictionary. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 453 | dco = zlib.decompressobj() |
| 454 | self.assertRaises(zlib.error, dco.decompress, cd) |
| 455 | |
| 456 | def test_dictionary_streaming(self): |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 457 | # This simulates the reuse of a compressor object for compressing |
| 458 | # several separate data streams. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 459 | co = zlib.compressobj(zdict=HAMLET_SCENE) |
| 460 | do = zlib.decompressobj(zdict=HAMLET_SCENE) |
| 461 | piece = HAMLET_SCENE[1000:1500] |
| 462 | d0 = co.compress(piece) + co.flush(zlib.Z_SYNC_FLUSH) |
| 463 | d1 = co.compress(piece[100:]) + co.flush(zlib.Z_SYNC_FLUSH) |
| 464 | d2 = co.compress(piece[:-100]) + co.flush(zlib.Z_SYNC_FLUSH) |
| 465 | self.assertEqual(do.decompress(d0), piece) |
| 466 | self.assertEqual(do.decompress(d1), piece[100:]) |
| 467 | self.assertEqual(do.decompress(d2), piece[:-100]) |
| 468 | |
Antoine Pitrou | c09c92f | 2010-05-11 23:36:40 +0000 | [diff] [blame] | 469 | def test_decompress_incomplete_stream(self): |
| 470 | # This is 'foo', deflated |
| 471 | x = b'x\x9cK\xcb\xcf\x07\x00\x02\x82\x01E' |
| 472 | # For the record |
| 473 | self.assertEqual(zlib.decompress(x), b'foo') |
| 474 | self.assertRaises(zlib.error, zlib.decompress, x[:-5]) |
| 475 | # Omitting the stream end works with decompressor objects |
| 476 | # (see issue #8672). |
| 477 | dco = zlib.decompressobj() |
| 478 | y = dco.decompress(x[:-5]) |
| 479 | y += dco.flush() |
| 480 | self.assertEqual(y, b'foo') |
| 481 | |
Nadeem Vawda | 1c38546 | 2011-08-13 15:22:40 +0200 | [diff] [blame] | 482 | def test_decompress_eof(self): |
| 483 | x = b'x\x9cK\xcb\xcf\x07\x00\x02\x82\x01E' # 'foo' |
| 484 | dco = zlib.decompressobj() |
| 485 | self.assertFalse(dco.eof) |
| 486 | dco.decompress(x[:-5]) |
| 487 | self.assertFalse(dco.eof) |
| 488 | dco.decompress(x[-5:]) |
| 489 | self.assertTrue(dco.eof) |
| 490 | dco.flush() |
| 491 | self.assertTrue(dco.eof) |
| 492 | |
| 493 | def test_decompress_eof_incomplete_stream(self): |
| 494 | x = b'x\x9cK\xcb\xcf\x07\x00\x02\x82\x01E' # 'foo' |
| 495 | dco = zlib.decompressobj() |
| 496 | self.assertFalse(dco.eof) |
| 497 | dco.decompress(x[:-5]) |
| 498 | self.assertFalse(dco.eof) |
| 499 | dco.flush() |
| 500 | self.assertFalse(dco.eof) |
| 501 | |
Nadeem Vawda | 3907994 | 2012-11-05 00:37:42 +0100 | [diff] [blame] | 502 | def test_decompress_unused_data(self): |
| 503 | # Repeated calls to decompress() after EOF should accumulate data in |
| 504 | # dco.unused_data, instead of just storing the arg to the last call. |
Nadeem Vawda | ee7889d | 2012-11-11 02:14:36 +0100 | [diff] [blame] | 505 | source = b'abcdefghijklmnopqrstuvwxyz' |
| 506 | remainder = b'0123456789' |
| 507 | y = zlib.compress(source) |
| 508 | x = y + remainder |
| 509 | for maxlen in 0, 1000: |
| 510 | for step in 1, 2, len(y), len(x): |
| 511 | dco = zlib.decompressobj() |
| 512 | data = b'' |
| 513 | for i in range(0, len(x), step): |
| 514 | if i < len(y): |
| 515 | self.assertEqual(dco.unused_data, b'') |
| 516 | if maxlen == 0: |
| 517 | data += dco.decompress(x[i : i + step]) |
| 518 | self.assertEqual(dco.unconsumed_tail, b'') |
| 519 | else: |
| 520 | data += dco.decompress( |
| 521 | dco.unconsumed_tail + x[i : i + step], maxlen) |
| 522 | data += dco.flush() |
Nadeem Vawda | dd1253a | 2012-11-11 02:21:22 +0100 | [diff] [blame] | 523 | self.assertTrue(dco.eof) |
Nadeem Vawda | ee7889d | 2012-11-11 02:14:36 +0100 | [diff] [blame] | 524 | self.assertEqual(data, source) |
| 525 | self.assertEqual(dco.unconsumed_tail, b'') |
| 526 | self.assertEqual(dco.unused_data, remainder) |
Nadeem Vawda | 3907994 | 2012-11-05 00:37:42 +0100 | [diff] [blame] | 527 | |
Nadeem Vawda | 7ee9555 | 2012-11-11 03:15:32 +0100 | [diff] [blame] | 528 | def test_flush_with_freed_input(self): |
| 529 | # Issue #16411: decompressor accesses input to last decompress() call |
| 530 | # in flush(), even if this object has been freed in the meanwhile. |
| 531 | input1 = b'abcdefghijklmnopqrstuvwxyz' |
| 532 | input2 = b'QWERTYUIOPASDFGHJKLZXCVBNM' |
| 533 | data = zlib.compress(input1) |
| 534 | dco = zlib.decompressobj() |
| 535 | dco.decompress(data, 1) |
| 536 | del data |
| 537 | data = zlib.compress(input2) |
| 538 | self.assertEqual(dco.flush(), input1[1:]) |
| 539 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 540 | @requires_Compress_copy |
| 541 | def test_compresscopy(self): |
| 542 | # Test copying a compression object |
| 543 | data0 = HAMLET_SCENE |
| 544 | data1 = bytes(str(HAMLET_SCENE, "ascii").swapcase(), "ascii") |
| 545 | c0 = zlib.compressobj(zlib.Z_BEST_COMPRESSION) |
| 546 | bufs0 = [] |
| 547 | bufs0.append(c0.compress(data0)) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 548 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 549 | c1 = c0.copy() |
| 550 | bufs1 = bufs0[:] |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 551 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 552 | bufs0.append(c0.compress(data0)) |
| 553 | bufs0.append(c0.flush()) |
| 554 | s0 = b''.join(bufs0) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 555 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 556 | bufs1.append(c1.compress(data1)) |
| 557 | bufs1.append(c1.flush()) |
| 558 | s1 = b''.join(bufs1) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 559 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 560 | self.assertEqual(zlib.decompress(s0),data0+data0) |
| 561 | self.assertEqual(zlib.decompress(s1),data0+data1) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 562 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 563 | @requires_Compress_copy |
| 564 | def test_badcompresscopy(self): |
| 565 | # Test copying a compression object in an inconsistent state |
| 566 | c = zlib.compressobj() |
| 567 | c.compress(HAMLET_SCENE) |
| 568 | c.flush() |
| 569 | self.assertRaises(ValueError, c.copy) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 570 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 571 | @requires_Decompress_copy |
| 572 | def test_decompresscopy(self): |
| 573 | # Test copying a decompression object |
| 574 | data = HAMLET_SCENE |
| 575 | comp = zlib.compress(data) |
| 576 | # Test type of return value |
| 577 | self.assertIsInstance(comp, bytes) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 578 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 579 | d0 = zlib.decompressobj() |
| 580 | bufs0 = [] |
| 581 | bufs0.append(d0.decompress(comp[:32])) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 582 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 583 | d1 = d0.copy() |
| 584 | bufs1 = bufs0[:] |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 585 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 586 | bufs0.append(d0.decompress(comp[32:])) |
| 587 | s0 = b''.join(bufs0) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 588 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 589 | bufs1.append(d1.decompress(comp[32:])) |
| 590 | s1 = b''.join(bufs1) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 591 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 592 | self.assertEqual(s0,s1) |
| 593 | self.assertEqual(s0,data) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 594 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 595 | @requires_Decompress_copy |
| 596 | def test_baddecompresscopy(self): |
| 597 | # Test copying a compression object in an inconsistent state |
| 598 | data = zlib.compress(HAMLET_SCENE) |
| 599 | d = zlib.decompressobj() |
| 600 | d.decompress(data) |
| 601 | d.flush() |
| 602 | self.assertRaises(ValueError, d.copy) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 603 | |
Serhiy Storchaka | d7a4415 | 2015-11-12 11:23:04 +0200 | [diff] [blame] | 604 | def test_compresspickle(self): |
| 605 | for proto in range(pickle.HIGHEST_PROTOCOL + 1): |
| 606 | with self.assertRaises((TypeError, pickle.PicklingError)): |
| 607 | pickle.dumps(zlib.compressobj(zlib.Z_BEST_COMPRESSION), proto) |
| 608 | |
| 609 | def test_decompresspickle(self): |
| 610 | for proto in range(pickle.HIGHEST_PROTOCOL + 1): |
| 611 | with self.assertRaises((TypeError, pickle.PicklingError)): |
| 612 | pickle.dumps(zlib.decompressobj(), proto) |
| 613 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 614 | # Memory use of the following functions takes into account overallocation |
| 615 | |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 616 | @bigmemtest(size=_1G + 1024 * 1024, memuse=3) |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 617 | def test_big_compress_buffer(self, size): |
| 618 | c = zlib.compressobj(1) |
| 619 | compress = lambda s: c.compress(s) + c.flush() |
| 620 | self.check_big_compress_buffer(size, compress) |
| 621 | |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 622 | @bigmemtest(size=_1G + 1024 * 1024, memuse=2) |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 623 | def test_big_decompress_buffer(self, size): |
| 624 | d = zlib.decompressobj() |
| 625 | decompress = lambda s: d.decompress(s) + d.flush() |
| 626 | self.check_big_decompress_buffer(size, decompress) |
| 627 | |
Nadeem Vawda | 197e22c | 2012-02-23 14:23:17 +0200 | [diff] [blame] | 628 | @bigmemtest(size=_4G + 100, memuse=1, dry_run=False) |
Nadeem Vawda | 0c3d96a | 2011-05-15 00:19:50 +0200 | [diff] [blame] | 629 | def test_length_overflow(self, size): |
Nadeem Vawda | 0c3d96a | 2011-05-15 00:19:50 +0200 | [diff] [blame] | 630 | data = b'x' * size |
Nadeem Vawda | 1161a9c | 2011-05-15 00:48:24 +0200 | [diff] [blame] | 631 | c = zlib.compressobj(1) |
| 632 | d = zlib.decompressobj() |
Nadeem Vawda | 0c3d96a | 2011-05-15 00:19:50 +0200 | [diff] [blame] | 633 | try: |
Nadeem Vawda | 1161a9c | 2011-05-15 00:48:24 +0200 | [diff] [blame] | 634 | self.assertRaises(OverflowError, c.compress, data) |
| 635 | self.assertRaises(OverflowError, d.decompress, data) |
Nadeem Vawda | 0c3d96a | 2011-05-15 00:19:50 +0200 | [diff] [blame] | 636 | finally: |
| 637 | data = None |
| 638 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 639 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 640 | def genblock(seed, length, step=1024, generator=random): |
| 641 | """length-byte stream of random data from a seed (in step-byte blocks).""" |
| 642 | if seed is not None: |
| 643 | generator.seed(seed) |
| 644 | randint = generator.randint |
| 645 | if length < step or step < 2: |
| 646 | step = length |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 647 | blocks = bytes() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 648 | for i in range(0, length, step): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 649 | blocks += bytes(randint(0, 255) for x in range(step)) |
| 650 | return blocks |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 651 | |
| 652 | |
| 653 | |
| 654 | def choose_lines(source, number, seed=None, generator=random): |
| 655 | """Return a list of number lines randomly chosen from the source""" |
| 656 | if seed is not None: |
| 657 | generator.seed(seed) |
| 658 | sources = source.split('\n') |
| 659 | return [generator.choice(sources) for n in range(number)] |
| 660 | |
| 661 | |
| 662 | |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 663 | HAMLET_SCENE = b""" |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 664 | LAERTES |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 665 | |
| 666 | O, fear me not. |
| 667 | I stay too long: but here my father comes. |
| 668 | |
| 669 | Enter POLONIUS |
| 670 | |
| 671 | A double blessing is a double grace, |
| 672 | Occasion smiles upon a second leave. |
| 673 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 674 | LORD POLONIUS |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 675 | |
| 676 | Yet here, Laertes! aboard, aboard, for shame! |
| 677 | The wind sits in the shoulder of your sail, |
| 678 | And you are stay'd for. There; my blessing with thee! |
| 679 | And these few precepts in thy memory |
| 680 | See thou character. Give thy thoughts no tongue, |
| 681 | Nor any unproportioned thought his act. |
| 682 | Be thou familiar, but by no means vulgar. |
| 683 | Those friends thou hast, and their adoption tried, |
| 684 | Grapple them to thy soul with hoops of steel; |
| 685 | But do not dull thy palm with entertainment |
| 686 | Of each new-hatch'd, unfledged comrade. Beware |
| 687 | Of entrance to a quarrel, but being in, |
| 688 | Bear't that the opposed may beware of thee. |
| 689 | Give every man thy ear, but few thy voice; |
| 690 | Take each man's censure, but reserve thy judgment. |
| 691 | Costly thy habit as thy purse can buy, |
| 692 | But not express'd in fancy; rich, not gaudy; |
| 693 | For the apparel oft proclaims the man, |
| 694 | And they in France of the best rank and station |
| 695 | Are of a most select and generous chief in that. |
| 696 | Neither a borrower nor a lender be; |
| 697 | For loan oft loses both itself and friend, |
| 698 | And borrowing dulls the edge of husbandry. |
| 699 | This above all: to thine ownself be true, |
| 700 | And it must follow, as the night the day, |
| 701 | Thou canst not then be false to any man. |
| 702 | Farewell: my blessing season this in thee! |
| 703 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 704 | LAERTES |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 705 | |
| 706 | Most humbly do I take my leave, my lord. |
| 707 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 708 | LORD POLONIUS |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 709 | |
| 710 | The time invites you; go; your servants tend. |
| 711 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 712 | LAERTES |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 713 | |
| 714 | Farewell, Ophelia; and remember well |
| 715 | What I have said to you. |
| 716 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 717 | OPHELIA |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 718 | |
| 719 | 'Tis in my memory lock'd, |
| 720 | And you yourself shall keep the key of it. |
| 721 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 722 | LAERTES |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 723 | |
| 724 | Farewell. |
| 725 | """ |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 726 | |
| 727 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 728 | if __name__ == "__main__": |
Zachary Ware | 38c707e | 2015-04-13 15:00:43 -0500 | [diff] [blame] | 729 | unittest.main() |