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 |
Andrew M. Kuchling | 9a0f98e | 2001-02-21 02:17:01 +0000 | [diff] [blame] | 4 | import random |
Antoine Pitrou | f3d2275 | 2011-02-21 18:09:00 +0000 | [diff] [blame] | 5 | import sys |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 6 | from test.support import bigmemtest, _1G, _4G |
Andrew M. Kuchling | 9a0f98e | 2001-02-21 02:17:01 +0000 | [diff] [blame] | 7 | |
R. David Murray | a21e4ca | 2009-03-31 23:16:50 +0000 | [diff] [blame] | 8 | zlib = support.import_module('zlib') |
| 9 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 10 | requires_Compress_copy = unittest.skipUnless( |
| 11 | hasattr(zlib.compressobj(), "copy"), |
| 12 | 'requires Compress.copy()') |
| 13 | requires_Decompress_copy = unittest.skipUnless( |
| 14 | hasattr(zlib.decompressobj(), "copy"), |
| 15 | 'requires Decompress.copy()') |
| 16 | |
Andrew M. Kuchling | 9a0f98e | 2001-02-21 02:17:01 +0000 | [diff] [blame] | 17 | |
Nadeem Vawda | 64d25dd | 2011-09-12 00:04:13 +0200 | [diff] [blame] | 18 | class VersionTestCase(unittest.TestCase): |
| 19 | |
| 20 | def test_library_version(self): |
Nadeem Vawda | 131c707 | 2012-01-25 23:16:50 +0200 | [diff] [blame] | 21 | # Test that the major version of the actual library in use matches the |
| 22 | # major version that we were compiled against. We can't guarantee that |
| 23 | # the minor versions will match (even on the machine on which the module |
| 24 | # was compiled), and the API is stable between minor versions, so |
Nadeem Vawda | d770fe4 | 2012-01-28 17:32:47 +0200 | [diff] [blame] | 25 | # testing only the major versions avoids spurious failures. |
Nadeem Vawda | 131c707 | 2012-01-25 23:16:50 +0200 | [diff] [blame] | 26 | self.assertEqual(zlib.ZLIB_RUNTIME_VERSION[0], zlib.ZLIB_VERSION[0]) |
Nadeem Vawda | 64d25dd | 2011-09-12 00:04:13 +0200 | [diff] [blame] | 27 | |
| 28 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 29 | class ChecksumTestCase(unittest.TestCase): |
| 30 | # checksum test cases |
| 31 | def test_crc32start(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 32 | self.assertEqual(zlib.crc32(b""), zlib.crc32(b"", 0)) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 33 | self.assertTrue(zlib.crc32(b"abc", 0xffffffff)) |
Andrew M. Kuchling | fcfc8d5 | 2001-08-10 15:50:11 +0000 | [diff] [blame] | 34 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 35 | def test_crc32empty(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 36 | self.assertEqual(zlib.crc32(b"", 0), 0) |
| 37 | self.assertEqual(zlib.crc32(b"", 1), 1) |
| 38 | self.assertEqual(zlib.crc32(b"", 432), 432) |
Andrew M. Kuchling | 9a0f98e | 2001-02-21 02:17:01 +0000 | [diff] [blame] | 39 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 40 | def test_adler32start(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 41 | self.assertEqual(zlib.adler32(b""), zlib.adler32(b"", 1)) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 42 | self.assertTrue(zlib.adler32(b"abc", 0xffffffff)) |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 43 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 44 | def test_adler32empty(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 45 | self.assertEqual(zlib.adler32(b"", 0), 0) |
| 46 | self.assertEqual(zlib.adler32(b"", 1), 1) |
| 47 | self.assertEqual(zlib.adler32(b"", 432), 432) |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 48 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 49 | def assertEqual32(self, seen, expected): |
| 50 | # 32-bit values masked -- checksums on 32- vs 64- bit machines |
| 51 | # This is important if bit 31 (0x08000000L) is set. |
Guido van Rossum | e2a383d | 2007-01-15 16:59:06 +0000 | [diff] [blame] | 52 | self.assertEqual(seen & 0x0FFFFFFFF, expected & 0x0FFFFFFFF) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 53 | |
| 54 | def test_penguins(self): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 55 | self.assertEqual32(zlib.crc32(b"penguin", 0), 0x0e5c1a120) |
| 56 | self.assertEqual32(zlib.crc32(b"penguin", 1), 0x43b6aa94) |
| 57 | self.assertEqual32(zlib.adler32(b"penguin", 0), 0x0bcf02f6) |
| 58 | self.assertEqual32(zlib.adler32(b"penguin", 1), 0x0bd602f7) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 59 | |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 60 | self.assertEqual(zlib.crc32(b"penguin"), zlib.crc32(b"penguin", 0)) |
| 61 | self.assertEqual(zlib.adler32(b"penguin"),zlib.adler32(b"penguin",1)) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 62 | |
Gregory P. Smith | ab0d8a1 | 2008-03-17 20:24:09 +0000 | [diff] [blame] | 63 | def test_crc32_adler32_unsigned(self): |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 64 | foo = b'abcdefghijklmnop' |
Gregory P. Smith | ab0d8a1 | 2008-03-17 20:24:09 +0000 | [diff] [blame] | 65 | # explicitly test signed behavior |
Gregory P. Smith | 2727503 | 2008-03-20 06:20:09 +0000 | [diff] [blame] | 66 | self.assertEqual(zlib.crc32(foo), 2486878355) |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 67 | self.assertEqual(zlib.crc32(b'spam'), 1138425661) |
Gregory P. Smith | ab0d8a1 | 2008-03-17 20:24:09 +0000 | [diff] [blame] | 68 | self.assertEqual(zlib.adler32(foo+foo), 3573550353) |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 69 | self.assertEqual(zlib.adler32(b'spam'), 72286642) |
Gregory P. Smith | ab0d8a1 | 2008-03-17 20:24:09 +0000 | [diff] [blame] | 70 | |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 71 | def test_same_as_binascii_crc32(self): |
Martin v. Löwis | 15b16a3 | 2008-12-02 06:00:15 +0000 | [diff] [blame] | 72 | foo = b'abcdefghijklmnop' |
Gregory P. Smith | 2727503 | 2008-03-20 06:20:09 +0000 | [diff] [blame] | 73 | crc = 2486878355 |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 74 | self.assertEqual(binascii.crc32(foo), crc) |
| 75 | self.assertEqual(zlib.crc32(foo), crc) |
Martin v. Löwis | 15b16a3 | 2008-12-02 06:00:15 +0000 | [diff] [blame] | 76 | self.assertEqual(binascii.crc32(b'spam'), zlib.crc32(b'spam')) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 77 | |
| 78 | |
Antoine Pitrou | f3d2275 | 2011-02-21 18:09:00 +0000 | [diff] [blame] | 79 | # Issue #10276 - check that inputs >=4GB are handled correctly. |
| 80 | class ChecksumBigBufferTestCase(unittest.TestCase): |
| 81 | |
Nadeem Vawda | bc8c817 | 2012-02-23 14:16:15 +0200 | [diff] [blame] | 82 | @bigmemtest(size=_4G + 4, memuse=1, dry_run=False) |
| 83 | def test_big_buffer(self, size): |
Nadeem Vawda | b063a48 | 2012-02-23 13:36:25 +0200 | [diff] [blame] | 84 | data = b"nyan" * (_1G + 1) |
| 85 | self.assertEqual(zlib.crc32(data), 1044521549) |
| 86 | self.assertEqual(zlib.adler32(data), 2256789997) |
Antoine Pitrou | f3d2275 | 2011-02-21 18:09:00 +0000 | [diff] [blame] | 87 | |
Christian Heimes | b186d00 | 2008-03-18 15:15:01 +0000 | [diff] [blame] | 88 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 89 | class ExceptionTestCase(unittest.TestCase): |
| 90 | # make sure we generate some expected errors |
Guido van Rossum | 8ce8a78 | 2007-11-01 19:42:39 +0000 | [diff] [blame] | 91 | def test_badlevel(self): |
| 92 | # specifying compression level out of range causes an error |
| 93 | # (but -1 is Z_DEFAULT_COMPRESSION and apparently the zlib |
| 94 | # accepts 0 too) |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 95 | self.assertRaises(zlib.error, zlib.compress, b'ERROR', 10) |
| 96 | |
| 97 | def test_badargs(self): |
| 98 | self.assertRaises(TypeError, zlib.adler32) |
| 99 | self.assertRaises(TypeError, zlib.crc32) |
| 100 | self.assertRaises(TypeError, zlib.compress) |
| 101 | self.assertRaises(TypeError, zlib.decompress) |
| 102 | for arg in (42, None, '', 'abc', (), []): |
| 103 | self.assertRaises(TypeError, zlib.adler32, arg) |
| 104 | self.assertRaises(TypeError, zlib.crc32, arg) |
| 105 | self.assertRaises(TypeError, zlib.compress, arg) |
| 106 | self.assertRaises(TypeError, zlib.decompress, arg) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 107 | |
| 108 | def test_badcompressobj(self): |
| 109 | # verify failure on building compress object with bad params |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 110 | self.assertRaises(ValueError, zlib.compressobj, 1, zlib.DEFLATED, 0) |
Guido van Rossum | 8ce8a78 | 2007-11-01 19:42:39 +0000 | [diff] [blame] | 111 | # specifying total bits too large causes an error |
| 112 | self.assertRaises(ValueError, |
| 113 | zlib.compressobj, 1, zlib.DEFLATED, zlib.MAX_WBITS + 1) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 114 | |
| 115 | def test_baddecompressobj(self): |
| 116 | # verify failure on building decompress object with bad params |
Antoine Pitrou | 90ee4df | 2010-04-06 17:23:13 +0000 | [diff] [blame] | 117 | self.assertRaises(ValueError, zlib.decompressobj, -1) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 118 | |
Christian Heimes | 5e69685 | 2008-04-09 08:37:03 +0000 | [diff] [blame] | 119 | def test_decompressobj_badflush(self): |
| 120 | # verify failure on calling decompressobj.flush with bad params |
| 121 | self.assertRaises(ValueError, zlib.decompressobj().flush, 0) |
| 122 | self.assertRaises(ValueError, zlib.decompressobj().flush, -1) |
| 123 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 124 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 125 | class BaseCompressTestCase(object): |
| 126 | def check_big_compress_buffer(self, size, compress_func): |
| 127 | _1M = 1024 * 1024 |
| 128 | fmt = "%%0%dx" % (2 * _1M) |
| 129 | # Generate 10MB worth of random, and expand it by repeating it. |
| 130 | # The assumption is that zlib's memory is not big enough to exploit |
| 131 | # such spread out redundancy. |
| 132 | data = b''.join([random.getrandbits(8 * _1M).to_bytes(_1M, 'little') |
| 133 | for i in range(10)]) |
| 134 | data = data * (size // len(data) + 1) |
| 135 | try: |
| 136 | compress_func(data) |
| 137 | finally: |
| 138 | # Release memory |
| 139 | data = None |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 140 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 141 | def check_big_decompress_buffer(self, size, decompress_func): |
| 142 | data = b'x' * size |
| 143 | try: |
| 144 | compressed = zlib.compress(data, 1) |
| 145 | finally: |
| 146 | # Release memory |
| 147 | data = None |
| 148 | data = decompress_func(compressed) |
| 149 | # Sanity check |
| 150 | try: |
| 151 | self.assertEqual(len(data), size) |
| 152 | self.assertEqual(len(data.strip(b'x')), 0) |
| 153 | finally: |
| 154 | data = None |
| 155 | |
| 156 | |
| 157 | class CompressTestCase(BaseCompressTestCase, unittest.TestCase): |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 158 | # Test compression in one go (whole message compression) |
| 159 | def test_speech(self): |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 160 | x = zlib.compress(HAMLET_SCENE) |
| 161 | self.assertEqual(zlib.decompress(x), HAMLET_SCENE) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 162 | |
| 163 | def test_speech128(self): |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 164 | # compress more data |
| 165 | data = HAMLET_SCENE * 128 |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 166 | x = zlib.compress(data) |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 167 | self.assertEqual(zlib.compress(bytearray(data)), x) |
| 168 | for ob in x, bytearray(x): |
| 169 | self.assertEqual(zlib.decompress(ob), data) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 170 | |
Antoine Pitrou | 53b2166 | 2010-05-11 23:46:02 +0000 | [diff] [blame] | 171 | def test_incomplete_stream(self): |
| 172 | # An useful error message is given |
| 173 | x = zlib.compress(HAMLET_SCENE) |
Ezio Melotti | ed3a7d2 | 2010-12-01 02:32:32 +0000 | [diff] [blame] | 174 | self.assertRaisesRegex(zlib.error, |
Antoine Pitrou | 53b2166 | 2010-05-11 23:46:02 +0000 | [diff] [blame] | 175 | "Error -5 while decompressing data: incomplete or truncated stream", |
| 176 | zlib.decompress, x[:-1]) |
| 177 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 178 | # Memory use of the following functions takes into account overallocation |
| 179 | |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 180 | @bigmemtest(size=_1G + 1024 * 1024, memuse=3) |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 181 | def test_big_compress_buffer(self, size): |
| 182 | compress = lambda s: zlib.compress(s, 1) |
| 183 | self.check_big_compress_buffer(size, compress) |
| 184 | |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 185 | @bigmemtest(size=_1G + 1024 * 1024, memuse=2) |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 186 | def test_big_decompress_buffer(self, size): |
| 187 | self.check_big_decompress_buffer(size, zlib.decompress) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 188 | |
Nadeem Vawda | 197e22c | 2012-02-23 14:23:17 +0200 | [diff] [blame] | 189 | @bigmemtest(size=_4G + 100, memuse=1, dry_run=False) |
Victor Stinner | 8848c7a | 2011-01-04 02:07:36 +0000 | [diff] [blame] | 190 | def test_length_overflow(self, size): |
Victor Stinner | 8848c7a | 2011-01-04 02:07:36 +0000 | [diff] [blame] | 191 | data = b'x' * size |
| 192 | try: |
| 193 | self.assertRaises(OverflowError, zlib.compress, data, 1) |
Nadeem Vawda | 154bdf9 | 2011-05-14 23:07:36 +0200 | [diff] [blame] | 194 | self.assertRaises(OverflowError, zlib.decompress, data) |
Victor Stinner | 8848c7a | 2011-01-04 02:07:36 +0000 | [diff] [blame] | 195 | finally: |
| 196 | data = None |
| 197 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 198 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 199 | class CompressObjectTestCase(BaseCompressTestCase, unittest.TestCase): |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 200 | # Test compression object |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 201 | def test_pair(self): |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 202 | # straightforward compress/decompress objects |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 203 | datasrc = HAMLET_SCENE * 128 |
| 204 | datazip = zlib.compress(datasrc) |
| 205 | # should compress both bytes and bytearray data |
| 206 | for data in (datasrc, bytearray(datasrc)): |
| 207 | co = zlib.compressobj() |
| 208 | x1 = co.compress(data) |
| 209 | x2 = co.flush() |
| 210 | self.assertRaises(zlib.error, co.flush) # second flush should not work |
| 211 | self.assertEqual(x1 + x2, datazip) |
| 212 | for v1, v2 in ((x1, x2), (bytearray(x1), bytearray(x2))): |
| 213 | dco = zlib.decompressobj() |
| 214 | y1 = dco.decompress(v1 + v2) |
| 215 | y2 = dco.flush() |
| 216 | self.assertEqual(data, y1 + y2) |
| 217 | self.assertIsInstance(dco.unconsumed_tail, bytes) |
| 218 | self.assertIsInstance(dco.unused_data, bytes) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 219 | |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 220 | def test_compressoptions(self): |
| 221 | # specify lots of options to compressobj() |
| 222 | level = 2 |
| 223 | method = zlib.DEFLATED |
| 224 | wbits = -12 |
| 225 | memlevel = 9 |
| 226 | strategy = zlib.Z_FILTERED |
| 227 | co = zlib.compressobj(level, method, wbits, memlevel, strategy) |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 228 | x1 = co.compress(HAMLET_SCENE) |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 229 | x2 = co.flush() |
| 230 | dco = zlib.decompressobj(wbits) |
| 231 | y1 = dco.decompress(x1 + x2) |
| 232 | y2 = dco.flush() |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 233 | self.assertEqual(HAMLET_SCENE, y1 + y2) |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 234 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 235 | def test_compressincremental(self): |
| 236 | # compress object in steps, decompress object as one-shot |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 237 | data = HAMLET_SCENE * 128 |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 238 | co = zlib.compressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 239 | bufs = [] |
| 240 | for i in range(0, len(data), 256): |
| 241 | bufs.append(co.compress(data[i:i+256])) |
| 242 | bufs.append(co.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 243 | combuf = b''.join(bufs) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 244 | |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 245 | dco = zlib.decompressobj() |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 246 | y1 = dco.decompress(b''.join(bufs)) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 247 | y2 = dco.flush() |
| 248 | self.assertEqual(data, y1 + y2) |
| 249 | |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 250 | 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] | 251 | # compress object in steps, decompress object in steps |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 252 | source = source or HAMLET_SCENE |
| 253 | data = source * 128 |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 254 | co = zlib.compressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 255 | bufs = [] |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 256 | for i in range(0, len(data), cx): |
| 257 | bufs.append(co.compress(data[i:i+cx])) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 258 | bufs.append(co.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 259 | combuf = b''.join(bufs) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 260 | |
Gregory P. Smith | 693fc46 | 2008-09-06 20:13:06 +0000 | [diff] [blame] | 261 | decombuf = zlib.decompress(combuf) |
| 262 | # Test type of return value |
Ezio Melotti | e961593 | 2010-01-24 19:26:24 +0000 | [diff] [blame] | 263 | self.assertIsInstance(decombuf, bytes) |
Gregory P. Smith | 693fc46 | 2008-09-06 20:13:06 +0000 | [diff] [blame] | 264 | |
| 265 | self.assertEqual(data, decombuf) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 266 | |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 267 | dco = zlib.decompressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 268 | bufs = [] |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 269 | for i in range(0, len(combuf), dcx): |
| 270 | bufs.append(dco.decompress(combuf[i:i+dcx])) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 271 | self.assertEqual(b'', dco.unconsumed_tail, ######## |
| 272 | "(A) uct should be b'': not %d long" % |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 273 | len(dco.unconsumed_tail)) |
Amaury Forgeot d'Arc | e43d33a | 2008-07-02 20:50:16 +0000 | [diff] [blame] | 274 | self.assertEqual(b'', dco.unused_data) |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 275 | if flush: |
| 276 | bufs.append(dco.flush()) |
| 277 | else: |
| 278 | while True: |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 279 | chunk = dco.decompress(b'') |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 280 | if chunk: |
| 281 | bufs.append(chunk) |
| 282 | else: |
| 283 | break |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 284 | self.assertEqual(b'', dco.unconsumed_tail, ######## |
| 285 | "(B) uct should be b'': not %d long" % |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 286 | len(dco.unconsumed_tail)) |
Amaury Forgeot d'Arc | e43d33a | 2008-07-02 20:50:16 +0000 | [diff] [blame] | 287 | self.assertEqual(b'', dco.unused_data) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 288 | self.assertEqual(data, b''.join(bufs)) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 289 | # Failure means: "decompressobj with init options failed" |
| 290 | |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 291 | def test_decompincflush(self): |
| 292 | self.test_decompinc(flush=True) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 293 | |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 294 | def test_decompimax(self, source=None, cx=256, dcx=64): |
| 295 | # compress in steps, decompress in length-restricted steps |
| 296 | source = source or HAMLET_SCENE |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 297 | # Check a decompression object with max_length specified |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 298 | data = source * 128 |
| 299 | co = zlib.compressobj() |
| 300 | bufs = [] |
| 301 | for i in range(0, len(data), cx): |
| 302 | bufs.append(co.compress(data[i:i+cx])) |
| 303 | bufs.append(co.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 304 | combuf = b''.join(bufs) |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 305 | self.assertEqual(data, zlib.decompress(combuf), |
| 306 | 'compressed data failure') |
| 307 | |
| 308 | dco = zlib.decompressobj() |
| 309 | bufs = [] |
| 310 | cb = combuf |
| 311 | while cb: |
| 312 | #max_length = 1 + len(cb)//10 |
| 313 | chunk = dco.decompress(cb, dcx) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 314 | self.assertFalse(len(chunk) > dcx, |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 315 | 'chunk too big (%d>%d)' % (len(chunk), dcx)) |
| 316 | bufs.append(chunk) |
| 317 | cb = dco.unconsumed_tail |
| 318 | bufs.append(dco.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 319 | self.assertEqual(data, b''.join(bufs), 'Wrong data retrieved') |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 320 | |
| 321 | def test_decompressmaxlen(self, flush=False): |
| 322 | # Check a decompression object with max_length specified |
| 323 | data = HAMLET_SCENE * 128 |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 324 | co = zlib.compressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 325 | bufs = [] |
| 326 | for i in range(0, len(data), 256): |
| 327 | bufs.append(co.compress(data[i:i+256])) |
| 328 | bufs.append(co.flush()) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 329 | combuf = b''.join(bufs) |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 330 | self.assertEqual(data, zlib.decompress(combuf), |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 331 | 'compressed data failure') |
| 332 | |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 333 | dco = zlib.decompressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 334 | bufs = [] |
| 335 | cb = combuf |
| 336 | while cb: |
Guido van Rossum | f359410 | 2003-02-27 18:39:18 +0000 | [diff] [blame] | 337 | max_length = 1 + len(cb)//10 |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 338 | chunk = dco.decompress(cb, max_length) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 339 | self.assertFalse(len(chunk) > max_length, |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 340 | 'chunk too big (%d>%d)' % (len(chunk),max_length)) |
| 341 | bufs.append(chunk) |
| 342 | cb = dco.unconsumed_tail |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 343 | if flush: |
| 344 | bufs.append(dco.flush()) |
| 345 | else: |
| 346 | while chunk: |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 347 | chunk = dco.decompress(b'', max_length) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 348 | self.assertFalse(len(chunk) > max_length, |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 349 | 'chunk too big (%d>%d)' % (len(chunk),max_length)) |
| 350 | bufs.append(chunk) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 351 | self.assertEqual(data, b''.join(bufs), 'Wrong data retrieved') |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 352 | |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 353 | def test_decompressmaxlenflush(self): |
| 354 | self.test_decompressmaxlen(flush=True) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 355 | |
| 356 | def test_maxlenmisc(self): |
| 357 | # Misc tests of max_length |
Neil Schemenauer | 94afd3e | 2004-06-05 19:02:52 +0000 | [diff] [blame] | 358 | dco = zlib.decompressobj() |
Antoine Pitrou | 77b338b | 2009-12-14 18:00:06 +0000 | [diff] [blame] | 359 | self.assertRaises(ValueError, dco.decompress, b"", -1) |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 360 | self.assertEqual(b'', dco.unconsumed_tail) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 361 | |
Nadeem Vawda | 7619e88 | 2011-05-14 14:05:20 +0200 | [diff] [blame] | 362 | def test_clear_unconsumed_tail(self): |
| 363 | # Issue #12050: calling decompress() without providing max_length |
| 364 | # should clear the unconsumed_tail attribute. |
| 365 | cdata = b"x\x9cKLJ\x06\x00\x02M\x01" # "abc" |
| 366 | dco = zlib.decompressobj() |
| 367 | ddata = dco.decompress(cdata, 1) |
| 368 | ddata += dco.decompress(dco.unconsumed_tail) |
| 369 | self.assertEqual(dco.unconsumed_tail, b"") |
| 370 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 371 | def test_flushes(self): |
| 372 | # Test flush() with the various options, using all the |
| 373 | # different levels in order to provide more variations. |
| 374 | sync_opt = ['Z_NO_FLUSH', 'Z_SYNC_FLUSH', 'Z_FULL_FLUSH'] |
| 375 | sync_opt = [getattr(zlib, opt) for opt in sync_opt |
| 376 | if hasattr(zlib, opt)] |
Neil Schemenauer | 6412b12 | 2004-06-05 19:34:28 +0000 | [diff] [blame] | 377 | data = HAMLET_SCENE * 8 |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 378 | |
| 379 | for sync in sync_opt: |
| 380 | for level in range(10): |
| 381 | obj = zlib.compressobj( level ) |
| 382 | a = obj.compress( data[:3000] ) |
| 383 | b = obj.flush( sync ) |
| 384 | c = obj.compress( data[3000:] ) |
| 385 | d = obj.flush() |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 386 | self.assertEqual(zlib.decompress(b''.join([a,b,c,d])), |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 387 | data, ("Decompress failed: flush " |
| 388 | "mode=%i, level=%i") % (sync, level)) |
| 389 | del obj |
| 390 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 391 | @unittest.skipUnless(hasattr(zlib, 'Z_SYNC_FLUSH'), |
| 392 | 'requires zlib.Z_SYNC_FLUSH') |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 393 | def test_odd_flush(self): |
| 394 | # Test for odd flushing bugs noted in 2.0, and hopefully fixed in 2.1 |
| 395 | import random |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 396 | # Testing on 17K of "random" data |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 397 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 398 | # Create compressor and decompressor objects |
| 399 | co = zlib.compressobj(zlib.Z_BEST_COMPRESSION) |
| 400 | dco = zlib.decompressobj() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 401 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 402 | # Try 17K of data |
| 403 | # generate random data stream |
| 404 | try: |
| 405 | # In 2.3 and later, WichmannHill is the RNG of the bug report |
| 406 | gen = random.WichmannHill() |
| 407 | except AttributeError: |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 408 | try: |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 409 | # 2.2 called it Random |
| 410 | gen = random.Random() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 411 | except AttributeError: |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 412 | # others might simply have a single RNG |
| 413 | gen = random |
| 414 | gen.seed(1) |
| 415 | data = genblock(1, 17 * 1024, generator=gen) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 416 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 417 | # compress, sync-flush, and decompress |
| 418 | first = co.compress(data) |
| 419 | second = co.flush(zlib.Z_SYNC_FLUSH) |
| 420 | expanded = dco.decompress(first + second) |
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 | # if decompressed data is different from the input data, choke. |
| 423 | self.assertEqual(expanded, data, "17K random source doesn't match") |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 424 | |
Andrew M. Kuchling | 3b585b3 | 2004-12-28 20:10:48 +0000 | [diff] [blame] | 425 | def test_empty_flush(self): |
| 426 | # Test that calling .flush() on unused objects works. |
| 427 | # (Bug #1083110 -- calling .flush() on decompress objects |
| 428 | # caused a core dump.) |
| 429 | |
| 430 | co = zlib.compressobj(zlib.Z_BEST_COMPRESSION) |
Benjamin Peterson | c9c0f20 | 2009-06-30 23:06:06 +0000 | [diff] [blame] | 431 | self.assertTrue(co.flush()) # Returns a zlib header |
Andrew M. Kuchling | 3b585b3 | 2004-12-28 20:10:48 +0000 | [diff] [blame] | 432 | dco = zlib.decompressobj() |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 433 | self.assertEqual(dco.flush(), b"") # Returns nothing |
Tim Peters | 5a9fb3c | 2005-01-07 16:01:32 +0000 | [diff] [blame] | 434 | |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 435 | def test_dictionary(self): |
| 436 | h = HAMLET_SCENE |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 437 | # Build a simulated dictionary out of the words in HAMLET. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 438 | words = h.split() |
| 439 | random.shuffle(words) |
| 440 | zdict = b''.join(words) |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 441 | # Use it to compress HAMLET. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 442 | co = zlib.compressobj(zdict=zdict) |
| 443 | cd = co.compress(h) + co.flush() |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 444 | # Verify that it will decompress with the dictionary. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 445 | dco = zlib.decompressobj(zdict=zdict) |
| 446 | self.assertEqual(dco.decompress(cd) + dco.flush(), h) |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 447 | # Verify that it fails when not given the dictionary. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 448 | dco = zlib.decompressobj() |
| 449 | self.assertRaises(zlib.error, dco.decompress, cd) |
| 450 | |
| 451 | def test_dictionary_streaming(self): |
Nadeem Vawda | cf5e1d8 | 2012-06-22 00:35:57 +0200 | [diff] [blame] | 452 | # This simulates the reuse of a compressor object for compressing |
| 453 | # several separate data streams. |
Nadeem Vawda | fd8a838 | 2012-06-21 02:13:12 +0200 | [diff] [blame] | 454 | co = zlib.compressobj(zdict=HAMLET_SCENE) |
| 455 | do = zlib.decompressobj(zdict=HAMLET_SCENE) |
| 456 | piece = HAMLET_SCENE[1000:1500] |
| 457 | d0 = co.compress(piece) + co.flush(zlib.Z_SYNC_FLUSH) |
| 458 | d1 = co.compress(piece[100:]) + co.flush(zlib.Z_SYNC_FLUSH) |
| 459 | d2 = co.compress(piece[:-100]) + co.flush(zlib.Z_SYNC_FLUSH) |
| 460 | self.assertEqual(do.decompress(d0), piece) |
| 461 | self.assertEqual(do.decompress(d1), piece[100:]) |
| 462 | self.assertEqual(do.decompress(d2), piece[:-100]) |
| 463 | |
Antoine Pitrou | c09c92f | 2010-05-11 23:36:40 +0000 | [diff] [blame] | 464 | def test_decompress_incomplete_stream(self): |
| 465 | # This is 'foo', deflated |
| 466 | x = b'x\x9cK\xcb\xcf\x07\x00\x02\x82\x01E' |
| 467 | # For the record |
| 468 | self.assertEqual(zlib.decompress(x), b'foo') |
| 469 | self.assertRaises(zlib.error, zlib.decompress, x[:-5]) |
| 470 | # Omitting the stream end works with decompressor objects |
| 471 | # (see issue #8672). |
| 472 | dco = zlib.decompressobj() |
| 473 | y = dco.decompress(x[:-5]) |
| 474 | y += dco.flush() |
| 475 | self.assertEqual(y, b'foo') |
| 476 | |
Nadeem Vawda | 1c38546 | 2011-08-13 15:22:40 +0200 | [diff] [blame] | 477 | def test_decompress_eof(self): |
| 478 | x = b'x\x9cK\xcb\xcf\x07\x00\x02\x82\x01E' # 'foo' |
| 479 | dco = zlib.decompressobj() |
| 480 | self.assertFalse(dco.eof) |
| 481 | dco.decompress(x[:-5]) |
| 482 | self.assertFalse(dco.eof) |
| 483 | dco.decompress(x[-5:]) |
| 484 | self.assertTrue(dco.eof) |
| 485 | dco.flush() |
| 486 | self.assertTrue(dco.eof) |
| 487 | |
| 488 | def test_decompress_eof_incomplete_stream(self): |
| 489 | x = b'x\x9cK\xcb\xcf\x07\x00\x02\x82\x01E' # 'foo' |
| 490 | dco = zlib.decompressobj() |
| 491 | self.assertFalse(dco.eof) |
| 492 | dco.decompress(x[:-5]) |
| 493 | self.assertFalse(dco.eof) |
| 494 | dco.flush() |
| 495 | self.assertFalse(dco.eof) |
| 496 | |
Nadeem Vawda | 3907994 | 2012-11-05 00:37:42 +0100 | [diff] [blame] | 497 | def test_decompress_unused_data(self): |
| 498 | # Repeated calls to decompress() after EOF should accumulate data in |
| 499 | # 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] | 500 | source = b'abcdefghijklmnopqrstuvwxyz' |
| 501 | remainder = b'0123456789' |
| 502 | y = zlib.compress(source) |
| 503 | x = y + remainder |
| 504 | for maxlen in 0, 1000: |
| 505 | for step in 1, 2, len(y), len(x): |
| 506 | dco = zlib.decompressobj() |
| 507 | data = b'' |
| 508 | for i in range(0, len(x), step): |
| 509 | if i < len(y): |
| 510 | self.assertEqual(dco.unused_data, b'') |
| 511 | if maxlen == 0: |
| 512 | data += dco.decompress(x[i : i + step]) |
| 513 | self.assertEqual(dco.unconsumed_tail, b'') |
| 514 | else: |
| 515 | data += dco.decompress( |
| 516 | dco.unconsumed_tail + x[i : i + step], maxlen) |
| 517 | data += dco.flush() |
Nadeem Vawda | dd1253a | 2012-11-11 02:21:22 +0100 | [diff] [blame] | 518 | self.assertTrue(dco.eof) |
Nadeem Vawda | ee7889d | 2012-11-11 02:14:36 +0100 | [diff] [blame] | 519 | self.assertEqual(data, source) |
| 520 | self.assertEqual(dco.unconsumed_tail, b'') |
| 521 | self.assertEqual(dco.unused_data, remainder) |
Nadeem Vawda | 3907994 | 2012-11-05 00:37:42 +0100 | [diff] [blame] | 522 | |
Nadeem Vawda | 7ee9555 | 2012-11-11 03:15:32 +0100 | [diff] [blame] | 523 | def test_flush_with_freed_input(self): |
| 524 | # Issue #16411: decompressor accesses input to last decompress() call |
| 525 | # in flush(), even if this object has been freed in the meanwhile. |
| 526 | input1 = b'abcdefghijklmnopqrstuvwxyz' |
| 527 | input2 = b'QWERTYUIOPASDFGHJKLZXCVBNM' |
| 528 | data = zlib.compress(input1) |
| 529 | dco = zlib.decompressobj() |
| 530 | dco.decompress(data, 1) |
| 531 | del data |
| 532 | data = zlib.compress(input2) |
| 533 | self.assertEqual(dco.flush(), input1[1:]) |
| 534 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 535 | @requires_Compress_copy |
| 536 | def test_compresscopy(self): |
| 537 | # Test copying a compression object |
| 538 | data0 = HAMLET_SCENE |
| 539 | data1 = bytes(str(HAMLET_SCENE, "ascii").swapcase(), "ascii") |
| 540 | c0 = zlib.compressobj(zlib.Z_BEST_COMPRESSION) |
| 541 | bufs0 = [] |
| 542 | bufs0.append(c0.compress(data0)) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 543 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 544 | c1 = c0.copy() |
| 545 | bufs1 = bufs0[:] |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 546 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 547 | bufs0.append(c0.compress(data0)) |
| 548 | bufs0.append(c0.flush()) |
| 549 | s0 = b''.join(bufs0) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 550 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 551 | bufs1.append(c1.compress(data1)) |
| 552 | bufs1.append(c1.flush()) |
| 553 | s1 = b''.join(bufs1) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 554 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 555 | self.assertEqual(zlib.decompress(s0),data0+data0) |
| 556 | self.assertEqual(zlib.decompress(s1),data0+data1) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 557 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 558 | @requires_Compress_copy |
| 559 | def test_badcompresscopy(self): |
| 560 | # Test copying a compression object in an inconsistent state |
| 561 | c = zlib.compressobj() |
| 562 | c.compress(HAMLET_SCENE) |
| 563 | c.flush() |
| 564 | self.assertRaises(ValueError, c.copy) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 565 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 566 | @requires_Decompress_copy |
| 567 | def test_decompresscopy(self): |
| 568 | # Test copying a decompression object |
| 569 | data = HAMLET_SCENE |
| 570 | comp = zlib.compress(data) |
| 571 | # Test type of return value |
| 572 | self.assertIsInstance(comp, bytes) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 573 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 574 | d0 = zlib.decompressobj() |
| 575 | bufs0 = [] |
| 576 | bufs0.append(d0.decompress(comp[:32])) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 577 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 578 | d1 = d0.copy() |
| 579 | bufs1 = bufs0[:] |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 580 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 581 | bufs0.append(d0.decompress(comp[32:])) |
| 582 | s0 = b''.join(bufs0) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 583 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 584 | bufs1.append(d1.decompress(comp[32:])) |
| 585 | s1 = b''.join(bufs1) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 586 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 587 | self.assertEqual(s0,s1) |
| 588 | self.assertEqual(s0,data) |
Thomas Wouters | 477c8d5 | 2006-05-27 19:21:47 +0000 | [diff] [blame] | 589 | |
Serhiy Storchaka | 4376763 | 2013-11-03 21:31:38 +0200 | [diff] [blame] | 590 | @requires_Decompress_copy |
| 591 | def test_baddecompresscopy(self): |
| 592 | # Test copying a compression object in an inconsistent state |
| 593 | data = zlib.compress(HAMLET_SCENE) |
| 594 | d = zlib.decompressobj() |
| 595 | d.decompress(data) |
| 596 | d.flush() |
| 597 | self.assertRaises(ValueError, d.copy) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 598 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 599 | # Memory use of the following functions takes into account overallocation |
| 600 | |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 601 | @bigmemtest(size=_1G + 1024 * 1024, memuse=3) |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 602 | def test_big_compress_buffer(self, size): |
| 603 | c = zlib.compressobj(1) |
| 604 | compress = lambda s: c.compress(s) + c.flush() |
| 605 | self.check_big_compress_buffer(size, compress) |
| 606 | |
Antoine Pitrou | 94190bb | 2011-10-04 10:22:36 +0200 | [diff] [blame] | 607 | @bigmemtest(size=_1G + 1024 * 1024, memuse=2) |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 608 | def test_big_decompress_buffer(self, size): |
| 609 | d = zlib.decompressobj() |
| 610 | decompress = lambda s: d.decompress(s) + d.flush() |
| 611 | self.check_big_decompress_buffer(size, decompress) |
| 612 | |
Nadeem Vawda | 197e22c | 2012-02-23 14:23:17 +0200 | [diff] [blame] | 613 | @bigmemtest(size=_4G + 100, memuse=1, dry_run=False) |
Nadeem Vawda | 0c3d96a | 2011-05-15 00:19:50 +0200 | [diff] [blame] | 614 | def test_length_overflow(self, size): |
Nadeem Vawda | 0c3d96a | 2011-05-15 00:19:50 +0200 | [diff] [blame] | 615 | data = b'x' * size |
Nadeem Vawda | 1161a9c | 2011-05-15 00:48:24 +0200 | [diff] [blame] | 616 | c = zlib.compressobj(1) |
| 617 | d = zlib.decompressobj() |
Nadeem Vawda | 0c3d96a | 2011-05-15 00:19:50 +0200 | [diff] [blame] | 618 | try: |
Nadeem Vawda | 1161a9c | 2011-05-15 00:48:24 +0200 | [diff] [blame] | 619 | self.assertRaises(OverflowError, c.compress, data) |
| 620 | self.assertRaises(OverflowError, d.decompress, data) |
Nadeem Vawda | 0c3d96a | 2011-05-15 00:19:50 +0200 | [diff] [blame] | 621 | finally: |
| 622 | data = None |
| 623 | |
Antoine Pitrou | 8956271 | 2010-05-07 17:04:02 +0000 | [diff] [blame] | 624 | |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 625 | def genblock(seed, length, step=1024, generator=random): |
| 626 | """length-byte stream of random data from a seed (in step-byte blocks).""" |
| 627 | if seed is not None: |
| 628 | generator.seed(seed) |
| 629 | randint = generator.randint |
| 630 | if length < step or step < 2: |
| 631 | step = length |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 632 | blocks = bytes() |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 633 | for i in range(0, length, step): |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 634 | blocks += bytes(randint(0, 255) for x in range(step)) |
| 635 | return blocks |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 636 | |
| 637 | |
| 638 | |
| 639 | def choose_lines(source, number, seed=None, generator=random): |
| 640 | """Return a list of number lines randomly chosen from the source""" |
| 641 | if seed is not None: |
| 642 | generator.seed(seed) |
| 643 | sources = source.split('\n') |
| 644 | return [generator.choice(sources) for n in range(number)] |
| 645 | |
| 646 | |
| 647 | |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 648 | HAMLET_SCENE = b""" |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 649 | LAERTES |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 650 | |
| 651 | O, fear me not. |
| 652 | I stay too long: but here my father comes. |
| 653 | |
| 654 | Enter POLONIUS |
| 655 | |
| 656 | A double blessing is a double grace, |
| 657 | Occasion smiles upon a second leave. |
| 658 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 659 | LORD POLONIUS |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 660 | |
| 661 | Yet here, Laertes! aboard, aboard, for shame! |
| 662 | The wind sits in the shoulder of your sail, |
| 663 | And you are stay'd for. There; my blessing with thee! |
| 664 | And these few precepts in thy memory |
| 665 | See thou character. Give thy thoughts no tongue, |
| 666 | Nor any unproportioned thought his act. |
| 667 | Be thou familiar, but by no means vulgar. |
| 668 | Those friends thou hast, and their adoption tried, |
| 669 | Grapple them to thy soul with hoops of steel; |
| 670 | But do not dull thy palm with entertainment |
| 671 | Of each new-hatch'd, unfledged comrade. Beware |
| 672 | Of entrance to a quarrel, but being in, |
| 673 | Bear't that the opposed may beware of thee. |
| 674 | Give every man thy ear, but few thy voice; |
| 675 | Take each man's censure, but reserve thy judgment. |
| 676 | Costly thy habit as thy purse can buy, |
| 677 | But not express'd in fancy; rich, not gaudy; |
| 678 | For the apparel oft proclaims the man, |
| 679 | And they in France of the best rank and station |
| 680 | Are of a most select and generous chief in that. |
| 681 | Neither a borrower nor a lender be; |
| 682 | For loan oft loses both itself and friend, |
| 683 | And borrowing dulls the edge of husbandry. |
| 684 | This above all: to thine ownself be true, |
| 685 | And it must follow, as the night the day, |
| 686 | Thou canst not then be false to any man. |
| 687 | Farewell: my blessing season this in thee! |
| 688 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 689 | LAERTES |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 690 | |
| 691 | Most humbly do I take my leave, my lord. |
| 692 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 693 | LORD POLONIUS |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 694 | |
| 695 | The time invites you; go; your servants tend. |
| 696 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 697 | LAERTES |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 698 | |
| 699 | Farewell, Ophelia; and remember well |
| 700 | What I have said to you. |
| 701 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 702 | OPHELIA |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 703 | |
| 704 | 'Tis in my memory lock'd, |
| 705 | And you yourself shall keep the key of it. |
| 706 | |
Fred Drake | 004d5e6 | 2000-10-23 17:22:08 +0000 | [diff] [blame] | 707 | LAERTES |
Jeremy Hylton | 6eb4b6a | 1997-08-15 15:59:43 +0000 | [diff] [blame] | 708 | |
| 709 | Farewell. |
| 710 | """ |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 711 | |
| 712 | |
| 713 | def test_main(): |
Benjamin Peterson | ee8712c | 2008-05-20 21:35:26 +0000 | [diff] [blame] | 714 | support.run_unittest( |
Nadeem Vawda | 64d25dd | 2011-09-12 00:04:13 +0200 | [diff] [blame] | 715 | VersionTestCase, |
Walter Dörwald | 21d3a32 | 2003-05-01 17:45:56 +0000 | [diff] [blame] | 716 | ChecksumTestCase, |
Antoine Pitrou | f3d2275 | 2011-02-21 18:09:00 +0000 | [diff] [blame] | 717 | ChecksumBigBufferTestCase, |
Walter Dörwald | 21d3a32 | 2003-05-01 17:45:56 +0000 | [diff] [blame] | 718 | ExceptionTestCase, |
| 719 | CompressTestCase, |
| 720 | CompressObjectTestCase |
| 721 | ) |
Guido van Rossum | 7d9ea50 | 2003-02-03 20:45:52 +0000 | [diff] [blame] | 722 | |
| 723 | if __name__ == "__main__": |
Guido van Rossum | 776152b | 2007-05-22 22:44:07 +0000 | [diff] [blame] | 724 | unittest.main() # XXX |
| 725 | ###test_main() |