More changes needed to make things work once bytes and str are truly divorced.

Lib/hashlib.py

@@ -6,9 +6,9 @@
 
 __doc__ = """hashlib module - A common interface to many hash functions.
 
-new(name, string='') - returns a new hash object implementing the
+new(name, data=b'') - returns a new hash object implementing the
-                       given hash function; initializing the hash
+                      given hash function; initializing the hash
-                       using the given string data.
+                      using the given binary data.
 
 Named constructor functions are also available, these are much faster
 than using new():
@@ -39,14 +39,14 @@
 
     >>> import hashlib
     >>> m = hashlib.md5()
-    >>> m.update("Nobody inspects")
+    >>> m.update(b"Nobody inspects")
-    >>> m.update(" the spammish repetition")
+    >>> m.update(b" the spammish repetition")
     >>> m.digest()
-    '\xbbd\x9c\x83\xdd\x1e\xa5\xc9\xd9\xde\xc9\xa1\x8d\xf0\xff\xe9'
+    b'\xbbd\x9c\x83\xdd\x1e\xa5\xc9\xd9\xde\xc9\xa1\x8d\xf0\xff\xe9'
 
 More condensed:
 
-    >>> hashlib.sha224("Nobody inspects the spammish repetition").hexdigest()
+    >>> hashlib.sha224(b"Nobody inspects the spammish repetition").hexdigest()
     'a4337bc45a8fc544c03f52dc550cd6e1e87021bc896588bd79e901e2'
 
 """
@@ -77,25 +77,25 @@ def __get_builtin_constructor(name):
     raise ValueError, "unsupported hash type"
 
 
-def __py_new(name, string=''):
+def __py_new(name, data=b''):
-    """new(name, string='') - Return a new hashing object using the named algorithm;
+    """new(name, data='') - Return a new hashing object using the named algorithm;
-    optionally initialized with a string.
+    optionally initialized with data (which must be bytes).
     """
-    return __get_builtin_constructor(name)(string)
+    return __get_builtin_constructor(name)(data)
 
 
-def __hash_new(name, string=''):
+def __hash_new(name, data=b''):
-    """new(name, string='') - Return a new hashing object using the named algorithm;
+    """new(name, data=b'') - Return a new hashing object using the named algorithm;
-    optionally initialized with a string.
+    optionally initialized with data (which must be bytes).
     """
     try:
-        return _hashlib.new(name, string)
+        return _hashlib.new(name, data)
     except ValueError:
         # If the _hashlib module (OpenSSL) doesn't support the named
         # hash, try using our builtin implementations.
         # This allows for SHA224/256 and SHA384/512 support even though
         # the OpenSSL library prior to 0.9.8 doesn't provide them.
-        return __get_builtin_constructor(name)(string)
+        return __get_builtin_constructor(name)(data)
 
 
 try:

Lib/test/test_file.py

@@ -21,7 +21,7 @@ def tearDown(self):
     def testWeakRefs(self):
         # verify weak references
         p = proxy(self.f)
-        p.write('teststring')
+        p.write(b'teststring')
         self.assertEquals(self.f.tell(), p.tell())
         self.f.close()
         self.f = None
@@ -36,7 +36,7 @@ def testAttributes(self):
 
     def testReadinto(self):
         # verify readinto
-        self.f.write('12')
+        self.f.write(b'12')
         self.f.close()
         a = array('b', b'x'*10)
         self.f = open(TESTFN, 'rb')
@@ -215,7 +215,7 @@ def testIteration(self):
         # Test the complex interaction when mixing file-iteration and the
         # various read* methods.
         dataoffset = 16384
-        filler = "ham\n"
+        filler = b"ham\n"
         assert not dataoffset % len(filler), \
             "dataoffset must be multiple of len(filler)"
         nchunks = dataoffset // len(filler)

Lib/test/test_fileinput.py

@@ -27,7 +27,8 @@ def writeTmp(i, lines, mode='w'):  # opening in text mode is the default
 
 def remove_tempfiles(*names):
     for name in names:
-        safe_unlink(name)
+        if name:
+            safe_unlink(name)
 
 class BufferSizesTests(unittest.TestCase):
     def test_buffer_sizes(self):
@@ -191,30 +192,33 @@ def test_opening_mode(self):
             self.fail("FileInput should reject invalid mode argument")
         except ValueError:
             pass
+        t1 = None
         try:
             # try opening in universal newline mode
-            t1 = writeTmp(1, ["A\nB\r\nC\rD"], mode="wb")
+            t1 = writeTmp(1, [b"A\nB\r\nC\rD"], mode="wb")
             fi = FileInput(files=t1, mode="U")
             lines = list(fi)
             self.assertEqual(lines, ["A\n", "B\n", "C\n", "D"])
         finally:
             remove_tempfiles(t1)
 
-##     def test_file_opening_hook(self):
+    def test_file_opening_hook(self):
-##         # XXX The rot13 codec was removed.
+        try:
-##         #     So this test needs to be changed to use something else.
+            # cannot use openhook and inplace mode
-##         try:
+            fi = FileInput(inplace=1, openhook=lambda f, m: None)
-##             # cannot use openhook and inplace mode
+            self.fail("FileInput should raise if both inplace "
-##             fi = FileInput(inplace=1, openhook=lambda f, m: None)
+                             "and openhook arguments are given")
-##             self.fail("FileInput should raise if both inplace "
+        except ValueError:
-##                              "and openhook arguments are given")
+            pass
-##         except ValueError:
+        try:
-##             pass
+            fi = FileInput(openhook=1)
-##         try:
+            self.fail("FileInput should check openhook for being callable")
-##             fi = FileInput(openhook=1)
+        except ValueError:
-##             self.fail("FileInput should check openhook for being callable")
+            pass
-##         except ValueError:
+        # XXX The rot13 codec was removed.
-##             pass
+        #     So this test needs to be changed to use something else.
+        #     (Or perhaps the API needs to change so we can just pass
+        #     an encoding rather than using a hook?)
 ##         try:
 ##             t1 = writeTmp(1, ["A\nB"], mode="wb")
 ##             fi = FileInput(files=t1, openhook=hook_encoded("rot13"))

Lib/test/test_hashlib.py

@@ -41,9 +41,9 @@ def test_hexdigest(self):
 
 
     def test_large_update(self):
-        aas = 'a' * 128
+        aas = b'a' * 128
-        bees = 'b' * 127
+        bees = b'b' * 127
-        cees = 'c' * 126
+        cees = b'c' * 126
 
         for name in self.supported_hash_names:
             m1 = hashlib.new(name)
@@ -104,83 +104,83 @@ def test_case_sha1_3(self):
     # http://csrc.nist.gov/publications/fips/fips180-2/fips180-2.pdf
 
     def test_case_sha224_0(self):
-        self.check('sha224', "",
+        self.check('sha224', b"",
           "d14a028c2a3a2bc9476102bb288234c415a2b01f828ea62ac5b3e42f")
 
     def test_case_sha224_1(self):
-        self.check('sha224', "abc",
+        self.check('sha224', b"abc",
           "23097d223405d8228642a477bda255b32aadbce4bda0b3f7e36c9da7")
 
     def test_case_sha224_2(self):
         self.check('sha224',
-          "abcdbcdecdefdefgefghfghighijhijkijkljklmklmnlmnomnopnopq",
+          b"abcdbcdecdefdefgefghfghighijhijkijkljklmklmnlmnomnopnopq",
           "75388b16512776cc5dba5da1fd890150b0c6455cb4f58b1952522525")
 
     def test_case_sha224_3(self):
-        self.check('sha224', "a" * 1000000,
+        self.check('sha224', b"a" * 1000000,
           "20794655980c91d8bbb4c1ea97618a4bf03f42581948b2ee4ee7ad67")
 
 
     def test_case_sha256_0(self):
-        self.check('sha256', "",
+        self.check('sha256', b"",
           "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855")
 
     def test_case_sha256_1(self):
-        self.check('sha256', "abc",
+        self.check('sha256', b"abc",
           "ba7816bf8f01cfea414140de5dae2223b00361a396177a9cb410ff61f20015ad")
 
     def test_case_sha256_2(self):
         self.check('sha256',
-          "abcdbcdecdefdefgefghfghighijhijkijkljklmklmnlmnomnopnopq",
+          b"abcdbcdecdefdefgefghfghighijhijkijkljklmklmnlmnomnopnopq",
           "248d6a61d20638b8e5c026930c3e6039a33ce45964ff2167f6ecedd419db06c1")
 
     def test_case_sha256_3(self):
-        self.check('sha256', "a" * 1000000,
+        self.check('sha256', b"a" * 1000000,
           "cdc76e5c9914fb9281a1c7e284d73e67f1809a48a497200e046d39ccc7112cd0")
 
 
     def test_case_sha384_0(self):
-        self.check('sha384', "",
+        self.check('sha384', b"",
           "38b060a751ac96384cd9327eb1b1e36a21fdb71114be07434c0cc7bf63f6e1da"+
           "274edebfe76f65fbd51ad2f14898b95b")
 
     def test_case_sha384_1(self):
-        self.check('sha384', "abc",
+        self.check('sha384', b"abc",
           "cb00753f45a35e8bb5a03d699ac65007272c32ab0eded1631a8b605a43ff5bed"+
           "8086072ba1e7cc2358baeca134c825a7")
 
     def test_case_sha384_2(self):
         self.check('sha384',
-                   "abcdefghbcdefghicdefghijdefghijkefghijklfghijklmghijklmn"+
+                   b"abcdefghbcdefghicdefghijdefghijkefghijklfghijklmghijklmn"+
-                   "hijklmnoijklmnopjklmnopqklmnopqrlmnopqrsmnopqrstnopqrstu",
+                   b"hijklmnoijklmnopjklmnopqklmnopqrlmnopqrsmnopqrstnopqrstu",
           "09330c33f71147e83d192fc782cd1b4753111b173b3b05d22fa08086e3b0f712"+
           "fcc7c71a557e2db966c3e9fa91746039")
 
     def test_case_sha384_3(self):
-        self.check('sha384', "a" * 1000000,
+        self.check('sha384', b"a" * 1000000,
           "9d0e1809716474cb086e834e310a4a1ced149e9c00f248527972cec5704c2a5b"+
           "07b8b3dc38ecc4ebae97ddd87f3d8985")
 
 
     def test_case_sha512_0(self):
-        self.check('sha512', "",
+        self.check('sha512', b"",
           "cf83e1357eefb8bdf1542850d66d8007d620e4050b5715dc83f4a921d36ce9ce"+
           "47d0d13c5d85f2b0ff8318d2877eec2f63b931bd47417a81a538327af927da3e")
 
     def test_case_sha512_1(self):
-        self.check('sha512', "abc",
+        self.check('sha512', b"abc",
           "ddaf35a193617abacc417349ae20413112e6fa4e89a97ea20a9eeee64b55d39a"+
           "2192992a274fc1a836ba3c23a3feebbd454d4423643ce80e2a9ac94fa54ca49f")
 
     def test_case_sha512_2(self):
         self.check('sha512',
-                   "abcdefghbcdefghicdefghijdefghijkefghijklfghijklmghijklmn"+
+                   b"abcdefghbcdefghicdefghijdefghijkefghijklfghijklmghijklmn"+
-                   "hijklmnoijklmnopjklmnopqklmnopqrlmnopqrsmnopqrstnopqrstu",
+                   b"hijklmnoijklmnopjklmnopqklmnopqrlmnopqrsmnopqrstnopqrstu",
           "8e959b75dae313da8cf4f72814fc143f8f7779c6eb9f7fa17299aeadb6889018"+
           "501d289e4900f7e4331b99dec4b5433ac7d329eeb6dd26545e96e55b874be909")
 
     def test_case_sha512_3(self):
-        self.check('sha512', "a" * 1000000,
+        self.check('sha512', b"a" * 1000000,
           "e718483d0ce769644e2e42c7bc15b4638e1f98b13b2044285632a803afa973eb"+
           "de0ff244877ea60a4cb0432ce577c31beb009c5c2c49aa2e4eadb217ad8cc09b")
 

Lib/test/test_largefile.py

@@ -39,7 +39,7 @@
         # 2**31 == 2147483648
         f.seek(2147483649)
         # Seeking is not enough of a test: you must write and flush, too!
-        f.write("x")
+        f.write(b"x")
         f.flush()
     except (IOError, OverflowError):
         f.close()
@@ -70,10 +70,10 @@ def expect(got_this, expect_this):
     print('create large file via seek (may be sparse file) ...')
 f = open(name, 'wb')
 try:
-    f.write('z')
+    f.write(b'z')
     f.seek(0)
     f.seek(size)
-    f.write('a')
+    f.write(b'a')
     f.flush()
     if test_support.verbose:
         print('check file size with os.fstat')

Lib/test/test_multibytecodec.py

@@ -114,27 +114,28 @@ def test_dbcs(self):
     def test_dbcs_keep_buffer(self):
         decoder = codecs.getincrementaldecoder('cp949')()
         self.assertEqual(decoder.decode(b'\xc6\xc4\xc0'), '\ud30c')
-        self.assertRaises(UnicodeDecodeError, decoder.decode, '', True)
+        self.assertRaises(UnicodeDecodeError, decoder.decode, b'', True)
         self.assertEqual(decoder.decode(b'\xcc'), '\uc774')
 
         self.assertEqual(decoder.decode(b'\xc6\xc4\xc0'), '\ud30c')
-        self.assertRaises(UnicodeDecodeError, decoder.decode, '\xcc\xbd', True)
+        self.assertRaises(UnicodeDecodeError, decoder.decode,
+                          b'\xcc\xbd', True)
         self.assertEqual(decoder.decode(b'\xcc'), '\uc774')
 
     def test_iso2022(self):
         decoder = codecs.getincrementaldecoder('iso2022-jp')()
-        ESC = '\x1b'
+        ESC = b'\x1b'
-        self.assertEqual(decoder.decode(ESC + '('), '')
+        self.assertEqual(decoder.decode(ESC + b'('), '')
-        self.assertEqual(decoder.decode('B', True), '')
+        self.assertEqual(decoder.decode(b'B', True), '')
-        self.assertEqual(decoder.decode(ESC + '$'), '')
+        self.assertEqual(decoder.decode(ESC + b'$'), '')
-        self.assertEqual(decoder.decode('B@$'), '\u4e16')
+        self.assertEqual(decoder.decode(b'B@$'), '\u4e16')
-        self.assertEqual(decoder.decode('@$@'), '\u4e16')
+        self.assertEqual(decoder.decode(b'@$@'), '\u4e16')
-        self.assertEqual(decoder.decode('$', True), '\u4e16')
+        self.assertEqual(decoder.decode(b'$', True), '\u4e16')
         self.assertEqual(decoder.reset(), None)
-        self.assertEqual(decoder.decode('@$'), '@$')
+        self.assertEqual(decoder.decode(b'@$'), '@$')
-        self.assertEqual(decoder.decode(ESC + '$'), '')
+        self.assertEqual(decoder.decode(ESC + b'$'), '')
-        self.assertRaises(UnicodeDecodeError, decoder.decode, '', True)
+        self.assertRaises(UnicodeDecodeError, decoder.decode, b'', True)
-        self.assertEqual(decoder.decode('B@$'), '\u4e16')
+        self.assertEqual(decoder.decode(b'B@$'), '\u4e16')
 
 class Test_StreamReader(unittest.TestCase):
     def test_bug1728403(self):
@@ -213,7 +214,7 @@ def test_streamwriter_strwrite(self):
 
 class Test_ISO2022(unittest.TestCase):
     def test_g2(self):
-        iso2022jp2 = '\x1b(B:hu4:unit\x1b.A\x1bNi de famille'
+        iso2022jp2 = b'\x1b(B:hu4:unit\x1b.A\x1bNi de famille'
         uni = ':hu4:unit\xe9 de famille'
         self.assertEqual(iso2022jp2.decode('iso2022-jp-2'), uni)
 

Modules/socketmodule.c

@@ -1147,7 +1147,7 @@ getsockaddrarg(PySocketSockObject *s, PyObject *args,
 		struct sockaddr_un* addr;
 		char *path;
 		int len;
-		if (!PyArg_Parse(args, "t#", &path, &len))
+		if (!PyArg_Parse(args, "s#", &path, &len))
 			return 0;
 
 		addr = (struct sockaddr_un*)addr_ret;