Issue #26257: Eliminate buffer_tests.py and fix ByteArrayAsStringTest

ByteArrayAsStringTest.fixtype() was converting test data to bytes, not byte-
array, therefore many of the test cases inherited in this class were not
actually being run on the bytearray type.

The tests in buffer_tests.py were redundant with methods in string_tests
.MixinStrUnicodeUserStringTest and string_tests.CommonTest. These methods are
now moved into string_tests.BaseTest, where they will also get run for bytes
and bytearray.

This change also moves test_additional_split(), test_additional_rsplit(), and
test_strip() from CommonTest to BaseTest, meaning these tests are now run for
bytes and bytearray. I plan to eliminate redundancies with existing tests in
test_bytes.py soon.

Lib/test/buffer_tests.py

@@ -1,176 +0,0 @@
-# Tests that work for bytearray objects.  Could be merged into string_tests.
-# See PEP 3137.
-
-class MixinBytesBufferCommonTests(object):
-    """Tests that work for bytearray objects.
-    See PEP 3137.
-    """
-
-    def marshal(self, x):
-        """Convert x into the appropriate type for these tests."""
-        raise RuntimeError('test class must provide a marshal method')
-
-    def test_islower(self):
-        self.assertFalse(self.marshal(b'').islower())
-        self.assertTrue(self.marshal(b'a').islower())
-        self.assertFalse(self.marshal(b'A').islower())
-        self.assertFalse(self.marshal(b'\n').islower())
-        self.assertTrue(self.marshal(b'abc').islower())
-        self.assertFalse(self.marshal(b'aBc').islower())
-        self.assertTrue(self.marshal(b'abc\n').islower())
-        self.assertRaises(TypeError, self.marshal(b'abc').islower, 42)
-
-    def test_isupper(self):
-        self.assertFalse(self.marshal(b'').isupper())
-        self.assertFalse(self.marshal(b'a').isupper())
-        self.assertTrue(self.marshal(b'A').isupper())
-        self.assertFalse(self.marshal(b'\n').isupper())
-        self.assertTrue(self.marshal(b'ABC').isupper())
-        self.assertFalse(self.marshal(b'AbC').isupper())
-        self.assertTrue(self.marshal(b'ABC\n').isupper())
-        self.assertRaises(TypeError, self.marshal(b'abc').isupper, 42)
-
-    def test_istitle(self):
-        self.assertFalse(self.marshal(b'').istitle())
-        self.assertFalse(self.marshal(b'a').istitle())
-        self.assertTrue(self.marshal(b'A').istitle())
-        self.assertFalse(self.marshal(b'\n').istitle())
-        self.assertTrue(self.marshal(b'A Titlecased Line').istitle())
-        self.assertTrue(self.marshal(b'A\nTitlecased Line').istitle())
-        self.assertTrue(self.marshal(b'A Titlecased, Line').istitle())
-        self.assertFalse(self.marshal(b'Not a capitalized String').istitle())
-        self.assertFalse(self.marshal(b'Not\ta Titlecase String').istitle())
-        self.assertFalse(self.marshal(b'Not--a Titlecase String').istitle())
-        self.assertFalse(self.marshal(b'NOT').istitle())
-        self.assertRaises(TypeError, self.marshal(b'abc').istitle, 42)
-
-    def test_isspace(self):
-        self.assertFalse(self.marshal(b'').isspace())
-        self.assertFalse(self.marshal(b'a').isspace())
-        self.assertTrue(self.marshal(b' ').isspace())
-        self.assertTrue(self.marshal(b'\t').isspace())
-        self.assertTrue(self.marshal(b'\r').isspace())
-        self.assertTrue(self.marshal(b'\n').isspace())
-        self.assertTrue(self.marshal(b' \t\r\n').isspace())
-        self.assertFalse(self.marshal(b' \t\r\na').isspace())
-        self.assertRaises(TypeError, self.marshal(b'abc').isspace, 42)
-
-    def test_isalpha(self):
-        self.assertFalse(self.marshal(b'').isalpha())
-        self.assertTrue(self.marshal(b'a').isalpha())
-        self.assertTrue(self.marshal(b'A').isalpha())
-        self.assertFalse(self.marshal(b'\n').isalpha())
-        self.assertTrue(self.marshal(b'abc').isalpha())
-        self.assertFalse(self.marshal(b'aBc123').isalpha())
-        self.assertFalse(self.marshal(b'abc\n').isalpha())
-        self.assertRaises(TypeError, self.marshal(b'abc').isalpha, 42)
-
-    def test_isalnum(self):
-        self.assertFalse(self.marshal(b'').isalnum())
-        self.assertTrue(self.marshal(b'a').isalnum())
-        self.assertTrue(self.marshal(b'A').isalnum())
-        self.assertFalse(self.marshal(b'\n').isalnum())
-        self.assertTrue(self.marshal(b'123abc456').isalnum())
-        self.assertTrue(self.marshal(b'a1b3c').isalnum())
-        self.assertFalse(self.marshal(b'aBc000 ').isalnum())
-        self.assertFalse(self.marshal(b'abc\n').isalnum())
-        self.assertRaises(TypeError, self.marshal(b'abc').isalnum, 42)
-
-    def test_isdigit(self):
-        self.assertFalse(self.marshal(b'').isdigit())
-        self.assertFalse(self.marshal(b'a').isdigit())
-        self.assertTrue(self.marshal(b'0').isdigit())
-        self.assertTrue(self.marshal(b'0123456789').isdigit())
-        self.assertFalse(self.marshal(b'0123456789a').isdigit())
-
-        self.assertRaises(TypeError, self.marshal(b'abc').isdigit, 42)
-
-    def test_capitalize(self):
-        self.assertEqual(b' hello ', self.marshal(b' hello ').capitalize())
-        self.assertEqual(b'Hello ', self.marshal(b'Hello ').capitalize())
-        self.assertEqual(b'Hello ', self.marshal(b'hello ').capitalize())
-        self.assertEqual(b'Aaaa', self.marshal(b'aaaa').capitalize())
-        self.assertEqual(b'Aaaa', self.marshal(b'AaAa').capitalize())
-
-        self.assertRaises(TypeError, self.marshal(b'hello').capitalize, 42)
-
-    def test_ljust(self):
-        self.assertEqual(b'abc       ', self.marshal(b'abc').ljust(10))
-        self.assertEqual(b'abc   ', self.marshal(b'abc').ljust(6))
-        self.assertEqual(b'abc', self.marshal(b'abc').ljust(3))
-        self.assertEqual(b'abc', self.marshal(b'abc').ljust(2))
-        self.assertEqual(b'abc*******', self.marshal(b'abc').ljust(10, b'*'))
-        self.assertRaises(TypeError, self.marshal(b'abc').ljust)
-
-    def test_rjust(self):
-        self.assertEqual(b'       abc', self.marshal(b'abc').rjust(10))
-        self.assertEqual(b'   abc', self.marshal(b'abc').rjust(6))
-        self.assertEqual(b'abc', self.marshal(b'abc').rjust(3))
-        self.assertEqual(b'abc', self.marshal(b'abc').rjust(2))
-        self.assertEqual(b'*******abc', self.marshal(b'abc').rjust(10, b'*'))
-        self.assertRaises(TypeError, self.marshal(b'abc').rjust)
-
-    def test_center(self):
-        self.assertEqual(b'   abc    ', self.marshal(b'abc').center(10))
-        self.assertEqual(b' abc  ', self.marshal(b'abc').center(6))
-        self.assertEqual(b'abc', self.marshal(b'abc').center(3))
-        self.assertEqual(b'abc', self.marshal(b'abc').center(2))
-        self.assertEqual(b'***abc****', self.marshal(b'abc').center(10, b'*'))
-        self.assertRaises(TypeError, self.marshal(b'abc').center)
-
-    def test_swapcase(self):
-        self.assertEqual(b'hEllO CoMPuTErS',
-            self.marshal(b'HeLLo cOmpUteRs').swapcase())
-
-        self.assertRaises(TypeError, self.marshal(b'hello').swapcase, 42)
-
-    def test_zfill(self):
-        self.assertEqual(b'123', self.marshal(b'123').zfill(2))
-        self.assertEqual(b'123', self.marshal(b'123').zfill(3))
-        self.assertEqual(b'0123', self.marshal(b'123').zfill(4))
-        self.assertEqual(b'+123', self.marshal(b'+123').zfill(3))
-        self.assertEqual(b'+123', self.marshal(b'+123').zfill(4))
-        self.assertEqual(b'+0123', self.marshal(b'+123').zfill(5))
-        self.assertEqual(b'-123', self.marshal(b'-123').zfill(3))
-        self.assertEqual(b'-123', self.marshal(b'-123').zfill(4))
-        self.assertEqual(b'-0123', self.marshal(b'-123').zfill(5))
-        self.assertEqual(b'000', self.marshal(b'').zfill(3))
-        self.assertEqual(b'34', self.marshal(b'34').zfill(1))
-        self.assertEqual(b'0034', self.marshal(b'34').zfill(4))
-
-        self.assertRaises(TypeError, self.marshal(b'123').zfill)
-
-    def test_title(self):
-        self.assertEqual(b' Hello ', self.marshal(b' hello ').title())
-        self.assertEqual(b'Hello ', self.marshal(b'hello ').title())
-        self.assertEqual(b'Hello ', self.marshal(b'Hello ').title())
-        self.assertEqual(b'Format This As Title String',
-                         self.marshal(b'fOrMaT thIs aS titLe String').title())
-        self.assertEqual(b'Format,This-As*Title;String',
-                         self.marshal(b'fOrMaT,thIs-aS*titLe;String').title())
-        self.assertEqual(b'Getint', self.marshal(b'getInt').title())
-        self.assertRaises(TypeError, self.marshal(b'hello').title, 42)
-
-    def test_splitlines(self):
-        self.assertEqual([b'abc', b'def', b'', b'ghi'],
-                         self.marshal(b'abc\ndef\n\rghi').splitlines())
-        self.assertEqual([b'abc', b'def', b'', b'ghi'],
-                         self.marshal(b'abc\ndef\n\r\nghi').splitlines())
-        self.assertEqual([b'abc', b'def', b'ghi'],
-                         self.marshal(b'abc\ndef\r\nghi').splitlines())
-        self.assertEqual([b'abc', b'def', b'ghi'],
-                         self.marshal(b'abc\ndef\r\nghi\n').splitlines())
-        self.assertEqual([b'abc', b'def', b'ghi', b''],
-                         self.marshal(b'abc\ndef\r\nghi\n\r').splitlines())
-        self.assertEqual([b'', b'abc', b'def', b'ghi', b''],
-                         self.marshal(b'\nabc\ndef\r\nghi\n\r').splitlines())
-        self.assertEqual([b'', b'abc', b'def', b'ghi', b''],
-                         self.marshal(b'\nabc\ndef\r\nghi\n\r').splitlines(False))
-        self.assertEqual([b'\n', b'abc\n', b'def\r\n', b'ghi\n', b'\r'],
-                         self.marshal(b'\nabc\ndef\r\nghi\n\r').splitlines(True))
-        self.assertEqual([b'', b'abc', b'def', b'ghi', b''],
-                         self.marshal(b'\nabc\ndef\r\nghi\n\r').splitlines(keepends=False))
-        self.assertEqual([b'\n', b'abc\n', b'def\r\n', b'ghi\n', b'\r'],
-                         self.marshal(b'\nabc\ndef\r\nghi\n\r').splitlines(keepends=True))
-
-        self.assertRaises(TypeError, self.marshal(b'abc').splitlines, 42, 42)

Lib/test/string_tests.py

@@ -51,6 +51,9 @@ def fixtype(self, obj):
         else:
             return obj
 
+    def test_fixtype(self):
+        self.assertIs(type(self.fixtype("123")), self.type2test)
+
     # check that obj.method(*args) returns result
     def checkequal(self, result, obj, methodname, *args, **kwargs):
         result = self.fixtype(result)
@@ -682,21 +685,6 @@ def test_replace_overflow(self):
         self.checkraises(OverflowError, A2_16, "replace", "A", A2_16)
         self.checkraises(OverflowError, A2_16, "replace", "AA", A2_16+A2_16)
 
-
-
-class CommonTest(BaseTest):
-    # This testcase contains tests that can be used in all
-    # stringlike classes. Currently this is str and UserString.
-
-    def test_hash(self):
-        # SF bug 1054139:  += optimization was not invalidating cached hash value
-        a = self.type2test('DNSSEC')
-        b = self.type2test('')
-        for c in a:
-            b += c
-            hash(b)
-        self.assertEqual(hash(a), hash(b))
-
     def test_capitalize(self):
         self.checkequal(' hello ', ' hello ', 'capitalize')
         self.checkequal('Hello ', 'Hello ','capitalize')
@@ -704,23 +692,6 @@ def test_capitalize(self):
         self.checkequal('Aaaa', 'aaaa', 'capitalize')
         self.checkequal('Aaaa', 'AaAa', 'capitalize')
 
-        # check that titlecased chars are lowered correctly
-        # \u1ffc is the titlecased char
-        self.checkequal('\u03a9\u0399\u1ff3\u1ff3\u1ff3',
-                        '\u1ff3\u1ff3\u1ffc\u1ffc', 'capitalize')
-        # check with cased non-letter chars
-        self.checkequal('\u24c5\u24e8\u24e3\u24d7\u24de\u24dd',
-                        '\u24c5\u24ce\u24c9\u24bd\u24c4\u24c3', 'capitalize')
-        self.checkequal('\u24c5\u24e8\u24e3\u24d7\u24de\u24dd',
-                        '\u24df\u24e8\u24e3\u24d7\u24de\u24dd', 'capitalize')
-        self.checkequal('\u2160\u2171\u2172',
-                        '\u2160\u2161\u2162', 'capitalize')
-        self.checkequal('\u2160\u2171\u2172',
-                        '\u2170\u2171\u2172', 'capitalize')
-        # check with Ll chars with no upper - nothing changes here
-        self.checkequal('\u019b\u1d00\u1d86\u0221\u1fb7',
-                        '\u019b\u1d00\u1d86\u0221\u1fb7', 'capitalize')
-
         self.checkraises(TypeError, 'hello', 'capitalize', 42)
 
     def test_additional_split(self):
@@ -854,10 +825,6 @@ def test_zfill(self):
 
         self.checkraises(TypeError, '123', 'zfill')
 
-class MixinStrUnicodeUserStringTest:
-    # additional tests that only work for
-    # stringlike objects, i.e. str, UserString
-
     def test_islower(self):
         self.checkequal(False, '', 'islower')
         self.checkequal(True, 'a', 'islower')
@@ -960,6 +927,43 @@ def test_splitlines(self):
 
         self.checkraises(TypeError, 'abc', 'splitlines', 42, 42)
 
+
+class CommonTest(BaseTest):
+    # This testcase contains tests that can be used in all
+    # stringlike classes. Currently this is str and UserString.
+
+    def test_hash(self):
+        # SF bug 1054139:  += optimization was not invalidating cached hash value
+        a = self.type2test('DNSSEC')
+        b = self.type2test('')
+        for c in a:
+            b += c
+            hash(b)
+        self.assertEqual(hash(a), hash(b))
+
+    def test_capitalize_nonascii(self):
+        # check that titlecased chars are lowered correctly
+        # \u1ffc is the titlecased char
+        self.checkequal('\u03a9\u0399\u1ff3\u1ff3\u1ff3',
+                        '\u1ff3\u1ff3\u1ffc\u1ffc', 'capitalize')
+        # check with cased non-letter chars
+        self.checkequal('\u24c5\u24e8\u24e3\u24d7\u24de\u24dd',
+                        '\u24c5\u24ce\u24c9\u24bd\u24c4\u24c3', 'capitalize')
+        self.checkequal('\u24c5\u24e8\u24e3\u24d7\u24de\u24dd',
+                        '\u24df\u24e8\u24e3\u24d7\u24de\u24dd', 'capitalize')
+        self.checkequal('\u2160\u2171\u2172',
+                        '\u2160\u2161\u2162', 'capitalize')
+        self.checkequal('\u2160\u2171\u2172',
+                        '\u2170\u2171\u2172', 'capitalize')
+        # check with Ll chars with no upper - nothing changes here
+        self.checkequal('\u019b\u1d00\u1d86\u0221\u1fb7',
+                        '\u019b\u1d00\u1d86\u0221\u1fb7', 'capitalize')
+
+
+class MixinStrUnicodeUserStringTest:
+    # additional tests that only work for
+    # stringlike objects, i.e. str, UserString
+
     def test_startswith(self):
         self.checkequal(True, 'hello', 'startswith', 'he')
         self.checkequal(True, 'hello', 'startswith', 'hello')

Lib/test/test_bytes.py

@@ -1,8 +1,7 @@
 """Unit tests for the bytes and bytearray types.
 
-XXX This is a mess.  Common tests should be moved to buffer_tests.py,
+XXX This is a mess.  Common tests should be unified with string_tests.py (and
-which itself ought to be unified with string_tests.py (and the latter
+the latter should be modernized).
-should be modernized).
 """
 
 import os
@@ -16,7 +15,6 @@
 
 import test.support
 import test.string_tests
-import test.buffer_tests
 import test.list_tests
 from test.support import bigaddrspacetest, MAX_Py_ssize_t
 
@@ -1480,8 +1478,7 @@ def bytes_warning():
     # the rest that make sense (the code can be cleaned up to use modern
     # unittest methods at the same time).
 
-class BytearrayPEP3137Test(unittest.TestCase,
+class BytearrayPEP3137Test(unittest.TestCase):
-                       test.buffer_tests.MixinBytesBufferCommonTests):
     def marshal(self, x):
         return bytearray(x)
 
@@ -1511,16 +1508,16 @@ def test_returns_new_copy(self):
 class FixedStringTest(test.string_tests.BaseTest):
     def fixtype(self, obj):
         if isinstance(obj, str):
-            return obj.encode("utf-8")
+            return self.type2test(obj.encode("utf-8"))
         return super().fixtype(obj)
 
+    contains_bytes = True
+
 class ByteArrayAsStringTest(FixedStringTest, unittest.TestCase):
     type2test = bytearray
-    contains_bytes = True
 
 class BytesAsStringTest(FixedStringTest, unittest.TestCase):
     type2test = bytes
-    contains_bytes = True
 
 
 class SubclassTest:

Lib/test/test_unicode.py

@@ -565,7 +565,7 @@ def test_fixup(s):
         self.assertTrue('\ud800\udc02' < '\ud84d\udc56')
 
     def test_islower(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_islower(self)
+        super().test_islower()
         self.checkequalnofix(False, '\u1FFc', 'islower')
         self.assertFalse('\u2167'.islower())
         self.assertTrue('\u2177'.islower())
@@ -580,7 +580,7 @@ def test_islower(self):
         self.assertFalse('\U0001F46F'.islower())
 
     def test_isupper(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_isupper(self)
+        super().test_isupper()
         if not sys.platform.startswith('java'):
             self.checkequalnofix(False, '\u1FFc', 'isupper')
         self.assertTrue('\u2167'.isupper())
@@ -596,7 +596,7 @@ def test_isupper(self):
         self.assertFalse('\U0001F46F'.isupper())
 
     def test_istitle(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_istitle(self)
+        super().test_istitle()
         self.checkequalnofix(True, '\u1FFc', 'istitle')
         self.checkequalnofix(True, 'Greek \u1FFcitlecases ...', 'istitle')
 
@@ -608,7 +608,7 @@ def test_istitle(self):
             self.assertFalse(ch.istitle(), '{!a} is not title'.format(ch))
 
     def test_isspace(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_isspace(self)
+        super().test_isspace()
         self.checkequalnofix(True, '\u2000', 'isspace')
         self.checkequalnofix(True, '\u200a', 'isspace')
         self.checkequalnofix(False, '\u2014', 'isspace')
@@ -618,13 +618,13 @@ def test_isspace(self):
             self.assertFalse(ch.isspace(), '{!a} is not space.'.format(ch))
 
     def test_isalnum(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_isalnum(self)
+        super().test_isalnum()
         for ch in ['\U00010401', '\U00010427', '\U00010429', '\U0001044E',
                    '\U0001D7F6', '\U00011066', '\U000104A0', '\U0001F107']:
             self.assertTrue(ch.isalnum(), '{!a} is alnum.'.format(ch))
 
     def test_isalpha(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_isalpha(self)
+        super().test_isalpha()
         self.checkequalnofix(True, '\u1FFc', 'isalpha')
         # non-BMP, cased
         self.assertTrue('\U00010401'.isalpha())
@@ -654,7 +654,7 @@ def test_isdecimal(self):
             self.assertTrue(ch.isdecimal(), '{!a} is decimal.'.format(ch))
 
     def test_isdigit(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_isdigit(self)
+        super().test_isdigit()
         self.checkequalnofix(True, '\u2460', 'isdigit')
         self.checkequalnofix(False, '\xbc', 'isdigit')
         self.checkequalnofix(True, '\u0660', 'isdigit')
@@ -807,7 +807,7 @@ def test_capitalize(self):
         self.assertEqual('A\u0345\u03a3'.capitalize(), 'A\u0345\u03c2')
 
     def test_title(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_title(self)
+        super().test_title()
         self.assertEqual('\U0001044F'.title(), '\U00010427')
         self.assertEqual('\U0001044F\U0001044F'.title(),
                          '\U00010427\U0001044F')