Update str related tests from 3.13.5 (#5953)

* Update str related tests from 3.13.5

* Apply RustPython patches

* Mark new failing tests

Lib/test/string_tests.py

@@ -8,18 +8,12 @@ from test.support import import_helper
 from collections import UserList
 import random
 
+
 class Sequence:
     def __init__(self, seq='wxyz'): self.seq = seq
     def __len__(self): return len(self.seq)
     def __getitem__(self, i): return self.seq[i]
 
-class BadSeq1(Sequence):
-    def __init__(self): self.seq = [7, 'hello', 123]
-    def __str__(self): return '{0} {1} {2}'.format(*self.seq)
-
-class BadSeq2(Sequence):
-    def __init__(self): self.seq = ['a', 'b', 'c']
-    def __len__(self): return 8
 
 class BaseTest:
     # These tests are for buffers of values (bytes) and not
@@ -27,7 +21,7 @@ class BaseTest:
     # and various string implementations
 
     # The type to be tested
-    # Change in subclasses to change the behaviour of fixtesttype()
+    # Change in subclasses to change the behaviour of fixtype()
     type2test = None
 
     # Whether the "contained items" of the container are integers in
@@ -36,7 +30,7 @@ class BaseTest:
     contains_bytes = False
 
     # All tests pass their arguments to the testing methods
-    # as str objects. fixtesttype() can be used to propagate
+    # as str objects. fixtype() can be used to propagate
     # these arguments to the appropriate type
     def fixtype(self, obj):
         if isinstance(obj, str):
@@ -160,6 +154,14 @@ class BaseTest:
                     self.assertEqual(rem, 0, '%s != 0 for %s' % (rem, i))
                     self.assertEqual(r1, r2, '%s != %s for %s' % (r1, r2, i))
 
+    # TODO: RUSTPYTHON; TypeError: Unexpected keyword argument count
+    @unittest.expectedFailure
+    def test_count_keyword(self):
+        self.assertEqual('aa'.replace('a', 'b', 0), 'aa'.replace('a', 'b', count=0))
+        self.assertEqual('aa'.replace('a', 'b', 1), 'aa'.replace('a', 'b', count=1))
+        self.assertEqual('aa'.replace('a', 'b', 2), 'aa'.replace('a', 'b', count=2))
+        self.assertEqual('aa'.replace('a', 'b', 3), 'aa'.replace('a', 'b', count=3))
+
     def test_find(self):
         self.checkequal(0, 'abcdefghiabc', 'find', 'abc')
         self.checkequal(9, 'abcdefghiabc', 'find', 'abc', 1)
@@ -327,11 +329,12 @@ class BaseTest:
             for i in range(len(s)):
                 if s.startswith(p, i):
                     return i
+            if p == '' and s == '':
+                return 0
             return -1
 
-        rr = random.randrange
-        choices = random.choices
-        for _ in range(1000):
+        def check_pattern(rr):
+            choices = random.choices
             p0 = ''.join(choices('abcde', k=rr(10))) * rr(10, 20)
             p = p0[:len(p0) - rr(10)] # pop off some characters
             left = ''.join(choices('abcdef', k=rr(2000)))
@@ -341,6 +344,49 @@ class BaseTest:
                 self.checkequal(reference_find(p, text),
                                 text, 'find', p)
 
+        rr = random.randrange
+        for _ in range(1000):
+            check_pattern(rr)
+
+        # Test that empty string always work:
+        check_pattern(lambda *args: 0)
+
+    def test_find_many_lengths(self):
+        haystack_repeats = [a * 10**e for e in range(6) for a in (1,2,5)]
+        haystacks = [(n, self.fixtype("abcab"*n + "da")) for n in haystack_repeats]
+
+        needle_repeats = [a * 10**e for e in range(6) for a in (1, 3)]
+        needles = [(m, self.fixtype("abcab"*m + "da")) for m in needle_repeats]
+
+        for n, haystack1 in haystacks:
+            haystack2 = haystack1[:-1]
+            for m, needle in needles:
+                answer1 = 5 * (n - m) if m <= n else -1
+                self.assertEqual(haystack1.find(needle), answer1, msg=(n,m))
+                self.assertEqual(haystack2.find(needle), -1, msg=(n,m))
+
+    def test_adaptive_find(self):
+        # This would be very slow for the naive algorithm,
+        # but str.find() should be O(n + m).
+        for N in 1000, 10_000, 100_000, 1_000_000:
+            A, B = 'a' * N, 'b' * N
+            haystack = A + A + B + A + A
+            needle = A + B + B + A
+            self.checkequal(-1, haystack, 'find', needle)
+            self.checkequal(0, haystack, 'count', needle)
+            self.checkequal(len(haystack), haystack + needle, 'find', needle)
+            self.checkequal(1, haystack + needle, 'count', needle)
+
+    def test_find_with_memory(self):
+        # Test the "Skip with memory" path in the two-way algorithm.
+        for N in 1000, 3000, 10_000, 30_000:
+            needle = 'ab' * N
+            haystack = ('ab'*(N-1) + 'b') * 2
+            self.checkequal(-1, haystack, 'find', needle)
+            self.checkequal(0, haystack, 'count', needle)
+            self.checkequal(len(haystack), haystack + needle, 'find', needle)
+            self.checkequal(1, haystack + needle, 'count', needle)
+
     def test_find_shift_table_overflow(self):
         """When the table of 8-bit shifts overflows."""
         N = 2**8 + 100
@@ -724,6 +770,18 @@ class BaseTest:
         self.checkraises(TypeError, 'hello', 'replace', 42, 'h')
         self.checkraises(TypeError, 'hello', 'replace', 'h', 42)
 
+    def test_replace_uses_two_way_maxcount(self):
+        # Test that maxcount works in _two_way_count in fastsearch.h
+        A, B = "A"*1000, "B"*1000
+        AABAA = A + A + B + A + A
+        ABBA = A + B + B + A
+        self.checkequal(AABAA + ABBA,
+                        AABAA + ABBA, 'replace', ABBA, "ccc", 0)
+        self.checkequal(AABAA + "ccc",
+                        AABAA + ABBA, 'replace', ABBA, "ccc", 1)
+        self.checkequal(AABAA + "ccc",
+                        AABAA + ABBA, 'replace', ABBA, "ccc", 2)
+
     @unittest.skip("TODO: RUSTPYTHON, may only apply to 32-bit platforms")
     @unittest.skipIf(sys.maxsize > (1 << 32) or struct.calcsize('P') != 4,
                      'only applies to 32-bit platforms')
@@ -734,8 +792,6 @@ class BaseTest:
         self.checkraises(OverflowError, A2_16, "replace", "A", A2_16)
         self.checkraises(OverflowError, A2_16, "replace", "AA", A2_16+A2_16)
 
-
-    # Python 3.9
     def test_removeprefix(self):
         self.checkequal('am', 'spam', 'removeprefix', 'sp')
         self.checkequal('spamspam', 'spamspamspam', 'removeprefix', 'spam')
@@ -754,7 +810,6 @@ class BaseTest:
         self.checkraises(TypeError, 'hello', 'removeprefix', 'h', 42)
         self.checkraises(TypeError, 'hello', 'removeprefix', ("he", "l"))
 
-    # Python 3.9
     def test_removesuffix(self):
         self.checkequal('sp', 'spam', 'removesuffix', 'am')
         self.checkequal('spamspam', 'spamspamspam', 'removesuffix', 'spam')
@@ -1053,7 +1108,7 @@ class BaseTest:
         self.checkraises(TypeError, 'abc', 'splitlines', 42, 42)
 
 
-class CommonTest(BaseTest):
+class StringLikeTest(BaseTest):
     # This testcase contains tests that can be used in all
     # stringlike classes. Currently this is str and UserString.
 
@@ -1084,11 +1139,6 @@ class CommonTest(BaseTest):
         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')
@@ -1273,8 +1323,11 @@ class MixinStrUnicodeUserStringTest:
             self.checkequal(((('a' * i) + '-') * i)[:-1], '-', 'join',
                  ('a' * i,) * i)
 
-        #self.checkequal(str(BadSeq1()), ' ', 'join', BadSeq1())
-        self.checkequal('a b c', ' ', 'join', BadSeq2())
+        class LiesAboutLengthSeq(Sequence):
+            def __init__(self): self.seq = ['a', 'b', 'c']
+            def __len__(self): return 8
+
+        self.checkequal('a b c', ' ', 'join', LiesAboutLengthSeq())
 
         self.checkraises(TypeError, ' ', 'join')
         self.checkraises(TypeError, ' ', 'join', None)
@@ -1459,19 +1512,19 @@ class MixinStrUnicodeUserStringTest:
         # issue 11828
         s = 'hello'
         x = 'x'
-        self.assertRaisesRegex(TypeError, r'^find\(', s.find,
+        self.assertRaisesRegex(TypeError, r'^find\b', s.find,
                                 x, None, None, None)
-        self.assertRaisesRegex(TypeError, r'^rfind\(', s.rfind,
+        self.assertRaisesRegex(TypeError, r'^rfind\b', s.rfind,
                                 x, None, None, None)
-        self.assertRaisesRegex(TypeError, r'^index\(', s.index,
+        self.assertRaisesRegex(TypeError, r'^index\b', s.index,
                                 x, None, None, None)
-        self.assertRaisesRegex(TypeError, r'^rindex\(', s.rindex,
+        self.assertRaisesRegex(TypeError, r'^rindex\b', s.rindex,
                                 x, None, None, None)
-        self.assertRaisesRegex(TypeError, r'^count\(', s.count,
+        self.assertRaisesRegex(TypeError, r'^count\b', s.count,
                                 x, None, None, None)
-        self.assertRaisesRegex(TypeError, r'^startswith\(', s.startswith,
+        self.assertRaisesRegex(TypeError, r'^startswith\b', s.startswith,
                                 x, None, None, None)
-        self.assertRaisesRegex(TypeError, r'^endswith\(', s.endswith,
+        self.assertRaisesRegex(TypeError, r'^endswith\b', s.endswith,
                                 x, None, None, None)
 
         # issue #15534

Lib/test/test_bytes.py

@@ -10,6 +10,7 @@ import re
 import sys
 import copy
 import functools
+import operator
 import pickle
 import tempfile
 import textwrap
@@ -46,6 +47,10 @@ class Indexable:
 
 class BaseBytesTest:
 
+    def assertTypedEqual(self, actual, expected):
+        self.assertIs(type(actual), type(expected))
+        self.assertEqual(actual, expected)
+
     def test_basics(self):
         b = self.type2test()
         self.assertEqual(type(b), self.type2test)
@@ -737,6 +742,37 @@ class BaseBytesTest:
         check(b'%i%b %*.*b', (10, b'3', 5, 3, b'abc',), b'103   abc')
         check(b'%c', b'a', b'a')
 
+        class PseudoFloat:
+            def __init__(self, value):
+                self.value = float(value)
+            def __int__(self):
+                return int(self.value)
+
+        pi = PseudoFloat(3.1415)
+
+        exceptions_params = [
+            ('%x format: an integer is required, not float', b'%x', 3.14),
+            ('%X format: an integer is required, not float', b'%X', 2.11),
+            ('%o format: an integer is required, not float', b'%o', 1.79),
+            ('%x format: an integer is required, not PseudoFloat', b'%x', pi),
+            ('%x format: an integer is required, not complex', b'%x', 3j),
+            ('%X format: an integer is required, not complex', b'%X', 2j),
+            ('%o format: an integer is required, not complex', b'%o', 1j),
+            ('%u format: a real number is required, not complex', b'%u', 3j),
+            # See https://github.com/python/cpython/issues/130928 as for why
+            # the exception message contains '%d' instead of '%i'.
+            ('%d format: a real number is required, not complex', b'%i', 2j),
+            ('%d format: a real number is required, not complex', b'%d', 2j),
+            (
+                r'%c requires an integer in range\(256\) or a single byte',
+                b'%c', pi
+            ),
+        ]
+
+        for msg, format_bytes, value in exceptions_params:
+            with self.assertRaisesRegex(TypeError, msg):
+                operator.mod(format_bytes, value)
+
     def test_imod(self):
         b = self.type2test(b'hello, %b!')
         orig = b
@@ -995,13 +1031,13 @@ class BaseBytesTest:
         self.assertEqual(c, b'hllo')
 
     def test_sq_item(self):
-        _testcapi = import_helper.import_module('_testcapi')
+        _testlimitedcapi = import_helper.import_module('_testlimitedcapi')
         obj = self.type2test((42,))
         with self.assertRaises(IndexError):
-            _testcapi.sequence_getitem(obj, -2)
+            _testlimitedcapi.sequence_getitem(obj, -2)
         with self.assertRaises(IndexError):
-            _testcapi.sequence_getitem(obj, 1)
-        self.assertEqual(_testcapi.sequence_getitem(obj, 0), 42)
+            _testlimitedcapi.sequence_getitem(obj, 1)
+        self.assertEqual(_testlimitedcapi.sequence_getitem(obj, 0), 42)
 
 
 class BytesTest(BaseBytesTest, unittest.TestCase):
@@ -1031,36 +1067,63 @@ class BytesTest(BaseBytesTest, unittest.TestCase):
             self.assertRaises(TypeError, f.readinto, b"")
 
     def test_custom(self):
-        class A:
-            def __bytes__(self):
-                return b'abc'
-        self.assertEqual(bytes(A()), b'abc')
-        class A: pass
-        self.assertRaises(TypeError, bytes, A())
-        class A:
-            def __bytes__(self):
-                return None
-        self.assertRaises(TypeError, bytes, A())
-        class A:
+        self.assertEqual(bytes(BytesSubclass(b'abc')), b'abc')
+        self.assertEqual(BytesSubclass(OtherBytesSubclass(b'abc')),
+                         BytesSubclass(b'abc'))
+        self.assertEqual(bytes(WithBytes(b'abc')), b'abc')
+        self.assertEqual(BytesSubclass(WithBytes(b'abc')), BytesSubclass(b'abc'))
+
+        class NoBytes: pass
+        self.assertRaises(TypeError, bytes, NoBytes())
+        self.assertRaises(TypeError, bytes, WithBytes('abc'))
+        self.assertRaises(TypeError, bytes, WithBytes(None))
+        class IndexWithBytes:
             def __bytes__(self):
                 return b'a'
             def __index__(self):
                 return 42
-        self.assertEqual(bytes(A()), b'a')
+        self.assertEqual(bytes(IndexWithBytes()), b'a')
         # Issue #25766
-        class A(str):
+        class StrWithBytes(str):
+            def __new__(cls, value):
+                self = str.__new__(cls, '\u20ac')
+                self.value = value
+                return self
             def __bytes__(self):
-                return b'abc'
-        self.assertEqual(bytes(A('\u20ac')), b'abc')
-        self.assertEqual(bytes(A('\u20ac'), 'iso8859-15'), b'\xa4')
+                return self.value
+        self.assertEqual(bytes(StrWithBytes(b'abc')), b'abc')
+        self.assertEqual(bytes(StrWithBytes(b'abc'), 'iso8859-15'), b'\xa4')
+        self.assertEqual(bytes(StrWithBytes(BytesSubclass(b'abc'))), b'abc')
+        self.assertEqual(BytesSubclass(StrWithBytes(b'abc')), BytesSubclass(b'abc'))
+        self.assertEqual(BytesSubclass(StrWithBytes(b'abc'), 'iso8859-15'),
+                         BytesSubclass(b'\xa4'))
+        self.assertEqual(BytesSubclass(StrWithBytes(BytesSubclass(b'abc'))),
+                         BytesSubclass(b'abc'))
+        self.assertEqual(BytesSubclass(StrWithBytes(OtherBytesSubclass(b'abc'))),
+                         BytesSubclass(b'abc'))
         # Issue #24731
-        class A:
+        self.assertTypedEqual(bytes(WithBytes(BytesSubclass(b'abc'))), BytesSubclass(b'abc'))
+        self.assertTypedEqual(BytesSubclass(WithBytes(BytesSubclass(b'abc'))),
+                              BytesSubclass(b'abc'))
+        self.assertTypedEqual(BytesSubclass(WithBytes(OtherBytesSubclass(b'abc'))),
+                              BytesSubclass(b'abc'))
+
+        class BytesWithBytes(bytes):
+            def __new__(cls, value):
+                self = bytes.__new__(cls, b'\xa4')
+                self.value = value
+                return self
             def __bytes__(self):
-                return OtherBytesSubclass(b'abc')
-        self.assertEqual(bytes(A()), b'abc')
-        self.assertIs(type(bytes(A())), OtherBytesSubclass)
-        self.assertEqual(BytesSubclass(A()), b'abc')
-        self.assertIs(type(BytesSubclass(A())), BytesSubclass)
+                return self.value
+        self.assertTypedEqual(bytes(BytesWithBytes(b'abc')), b'abc')
+        self.assertTypedEqual(BytesSubclass(BytesWithBytes(b'abc')),
+                              BytesSubclass(b'abc'))
+        self.assertTypedEqual(bytes(BytesWithBytes(BytesSubclass(b'abc'))),
+                              BytesSubclass(b'abc'))
+        self.assertTypedEqual(BytesSubclass(BytesWithBytes(BytesSubclass(b'abc'))),
+                              BytesSubclass(b'abc'))
+        self.assertTypedEqual(BytesSubclass(BytesWithBytes(OtherBytesSubclass(b'abc'))),
+                              BytesSubclass(b'abc'))
 
     # Test PyBytes_FromFormat()
     def test_from_format(self):
@@ -1233,6 +1296,8 @@ class BytesTest(BaseBytesTest, unittest.TestCase):
 class ByteArrayTest(BaseBytesTest, unittest.TestCase):
     type2test = bytearray
 
+    _testlimitedcapi = import_helper.import_module('_testlimitedcapi')
+
     def test_getitem_error(self):
         b = bytearray(b'python')
         msg = "bytearray indices must be integers or slices"
@@ -1325,47 +1390,73 @@ class ByteArrayTest(BaseBytesTest, unittest.TestCase):
         self.assertEqual(re.findall(br"\w+", b), [by("Hello"), by("world")])
 
     def test_setitem(self):
-        b = bytearray([1, 2, 3])
-        b[1] = 100
-        self.assertEqual(b, bytearray([1, 100, 3]))
-        b[-1] = 200
-        self.assertEqual(b, bytearray([1, 100, 200]))
-        b[0] = Indexable(10)
-        self.assertEqual(b, bytearray([10, 100, 200]))
-        try:
-            b[3] = 0
-            self.fail("Didn't raise IndexError")
-        except IndexError:
-            pass
-        try:
-            b[-10] = 0
-            self.fail("Didn't raise IndexError")
-        except IndexError:
-            pass
-        try:
-            b[0] = 256
-            self.fail("Didn't raise ValueError")
-        except ValueError:
-            pass
-        try:
-            b[0] = Indexable(-1)
-            self.fail("Didn't raise ValueError")
-        except ValueError:
-            pass
-        try:
-            b[0] = None
-            self.fail("Didn't raise TypeError")
-        except TypeError:
-            pass
+        def setitem_as_mapping(b, i, val):
+            b[i] = val
+
+        def setitem_as_sequence(b, i, val):
+            self._testlimitedcapi.sequence_setitem(b, i, val)
+
+        def do_tests(setitem):
+            b = bytearray([1, 2, 3])
+            setitem(b, 1, 100)
+            self.assertEqual(b, bytearray([1, 100, 3]))
+            setitem(b, -1, 200)
+            self.assertEqual(b, bytearray([1, 100, 200]))
+            setitem(b, 0, Indexable(10))
+            self.assertEqual(b, bytearray([10, 100, 200]))
+            try:
+                setitem(b, 3, 0)
+                self.fail("Didn't raise IndexError")
+            except IndexError:
+                pass
+            try:
+                setitem(b, -10, 0)
+                self.fail("Didn't raise IndexError")
+            except IndexError:
+                pass
+            try:
+                setitem(b, 0, 256)
+                self.fail("Didn't raise ValueError")
+            except ValueError:
+                pass
+            try:
+                setitem(b, 0, Indexable(-1))
+                self.fail("Didn't raise ValueError")
+            except ValueError:
+                pass
+            try:
+                setitem(b, 0, object())
+                self.fail("Didn't raise TypeError")
+            except TypeError:
+                pass
+
+        with self.subTest("tp_as_mapping"):
+            do_tests(setitem_as_mapping)
+
+        with self.subTest("tp_as_sequence"):
+            do_tests(setitem_as_sequence)
 
     def test_delitem(self):
-        b = bytearray(range(10))
-        del b[0]
-        self.assertEqual(b, bytearray(range(1, 10)))
-        del b[-1]
-        self.assertEqual(b, bytearray(range(1, 9)))
-        del b[4]
-        self.assertEqual(b, bytearray([1, 2, 3, 4, 6, 7, 8]))
+        def del_as_mapping(b, i):
+            del b[i]
+
+        def del_as_sequence(b, i):
+            self._testlimitedcapi.sequence_delitem(b, i)
+
+        def do_tests(delete):
+            b = bytearray(range(10))
+            delete(b, 0)
+            self.assertEqual(b, bytearray(range(1, 10)))
+            delete(b, -1)
+            self.assertEqual(b, bytearray(range(1, 9)))
+            delete(b, 4)
+            self.assertEqual(b, bytearray([1, 2, 3, 4, 6, 7, 8]))
+
+        with self.subTest("tp_as_mapping"):
+            do_tests(del_as_mapping)
+
+        with self.subTest("tp_as_sequence"):
+            do_tests(del_as_sequence)
 
     def test_setslice(self):
         b = bytearray(range(10))
@@ -1558,6 +1649,13 @@ class ByteArrayTest(BaseBytesTest, unittest.TestCase):
         a = bytearray(b'')
         a.extend([Indexable(ord('a'))])
         self.assertEqual(a, b'a')
+        a = bytearray(b'abc')
+        self.assertRaisesRegex(TypeError,  # Override for string.
+                               "expected iterable of integers; got: 'str'",
+                               a.extend, 'def')
+        self.assertRaisesRegex(TypeError,  # But not for others.
+                               "can't extend bytearray with float",
+                               a.extend, 1.0)
 
     def test_remove(self):
         b = bytearray(b'hello')
@@ -1747,6 +1845,8 @@ class ByteArrayTest(BaseBytesTest, unittest.TestCase):
         self.assertEqual(b3, b'xcxcxc')
 
     def test_mutating_index(self):
+        # See gh-91153
+
         class Boom:
             def __index__(self):
                 b.clear()
@@ -1758,10 +1858,9 @@ class ByteArrayTest(BaseBytesTest, unittest.TestCase):
                 b[0] = Boom()
 
         with self.subTest("tp_as_sequence"):
-            _testcapi = import_helper.import_module('_testcapi')
             b = bytearray(b'Now you see me...')
             with self.assertRaises(IndexError):
-                _testcapi.sequence_setitem(b, 0, Boom())
+                self._testlimitedcapi.sequence_setitem(b, 0, Boom())
 
 
 class AssortedBytesTest(unittest.TestCase):
@@ -2060,6 +2159,12 @@ class BytesSubclass(bytes):
 class OtherBytesSubclass(bytes):
     pass
 
+class WithBytes:
+    def __init__(self, value):
+        self.value = value
+    def __bytes__(self):
+        return self.value
+
 class ByteArraySubclassTest(SubclassTest, unittest.TestCase):
     basetype = bytearray
     type2test = ByteArraySubclass

Lib/test/test_str.py

@@ -7,6 +7,7 @@ Written by Marc-Andre Lemburg (mal@lemburg.com).
 """
 import _string
 import codecs
+import datetime
 import itertools
 import operator
 import pickle
@@ -55,8 +56,22 @@ def duplicate_string(text):
 class StrSubclass(str):
     pass
 
-class UnicodeTest(string_tests.CommonTest,
-        string_tests.MixinStrUnicodeUserStringTest,
+class OtherStrSubclass(str):
+    pass
+
+class WithStr:
+    def __init__(self, value):
+        self.value = value
+    def __str__(self):
+        return self.value
+
+class WithRepr:
+    def __init__(self, value):
+        self.value = value
+    def __repr__(self):
+        return self.value
+
+class StrTest(string_tests.StringLikeTest,
         string_tests.MixinStrUnicodeTest,
         unittest.TestCase):
 
@@ -84,6 +99,10 @@ class UnicodeTest(string_tests.CommonTest,
             self.assertEqual(realresult, result)
             self.assertTrue(object is not realresult)
 
+    def assertTypedEqual(self, actual, expected):
+        self.assertIs(type(actual), type(expected))
+        self.assertEqual(actual, expected)
+
     def test_literals(self):
         self.assertEqual('\xff', '\u00ff')
         self.assertEqual('\uffff', '\U0000ffff')
@@ -93,6 +112,8 @@ class UnicodeTest(string_tests.CommonTest,
         # raw strings should not have unicode escapes
         self.assertNotEqual(r"\u0020", " ")
 
+    # TODO: RUSTPYTHON
+    @unittest.expectedFailure
     def test_ascii(self):
         self.assertEqual(ascii('abc'), "'abc'")
         self.assertEqual(ascii('ab\\c'), "'ab\\\\c'")
@@ -128,10 +149,13 @@ class UnicodeTest(string_tests.CommonTest,
         self.assertEqual(ascii("\U00010000" * 39 + "\uffff" * 4096),
                             ascii("\U00010000" * 39 + "\uffff" * 4096))
 
-        class WrongRepr:
-            def __repr__(self):
-                return b'byte-repr'
-        self.assertRaises(TypeError, ascii, WrongRepr())
+        self.assertTypedEqual(ascii('\U0001f40d'), r"'\U0001f40d'")
+        self.assertTypedEqual(ascii(StrSubclass('abc')), "'abc'")
+        self.assertTypedEqual(ascii(WithRepr('<abc>')), '<abc>')
+        self.assertTypedEqual(ascii(WithRepr(StrSubclass('<abc>'))), StrSubclass('<abc>'))
+        self.assertTypedEqual(ascii(WithRepr('<\U0001f40d>')), r'<\U0001f40d>')
+        self.assertTypedEqual(ascii(WithRepr(StrSubclass('<\U0001f40d>'))), r'<\U0001f40d>')
+        self.assertRaises(TypeError, ascii, WithRepr(b'byte-repr'))
 
     def test_repr(self):
         # Test basic sanity of repr()
@@ -169,10 +193,13 @@ class UnicodeTest(string_tests.CommonTest,
         self.assertEqual(repr("\U00010000" * 39 + "\uffff" * 4096),
                             repr("\U00010000" * 39 + "\uffff" * 4096))
 
-        class WrongRepr:
-            def __repr__(self):
-                return b'byte-repr'
-        self.assertRaises(TypeError, repr, WrongRepr())
+        self.assertTypedEqual(repr('\U0001f40d'), "'\U0001f40d'")
+        self.assertTypedEqual(repr(StrSubclass('abc')), "'abc'")
+        self.assertTypedEqual(repr(WithRepr('<abc>')), '<abc>')
+        self.assertTypedEqual(repr(WithRepr(StrSubclass('<abc>'))), StrSubclass('<abc>'))
+        self.assertTypedEqual(repr(WithRepr('<\U0001f40d>')), '<\U0001f40d>')
+        self.assertTypedEqual(repr(WithRepr(StrSubclass('<\U0001f40d>'))), StrSubclass('<\U0001f40d>'))
+        self.assertRaises(TypeError, repr, WithRepr(b'byte-repr'))
 
     def test_iterators(self):
         # Make sure unicode objects have an __iter__ method
@@ -213,7 +240,7 @@ class UnicodeTest(string_tests.CommonTest,
                         self.assertEqual(case, pickled)
 
     def test_count(self):
-        string_tests.CommonTest.test_count(self)
+        string_tests.StringLikeTest.test_count(self)
         # check mixed argument types
         self.checkequalnofix(3,  'aaa', 'count', 'a')
         self.checkequalnofix(0,  'aaa', 'count', 'b')
@@ -243,7 +270,7 @@ class UnicodeTest(string_tests.CommonTest,
         self.checkequal(3, MyStr('aaa'), 'count', 'a')
 
     def test_find(self):
-        string_tests.CommonTest.test_find(self)
+        string_tests.StringLikeTest.test_find(self)
         # test implementation details of the memchr fast path
         self.checkequal(100, 'a' * 100 + '\u0102', 'find', '\u0102')
         self.checkequal(-1, 'a' * 100 + '\u0102', 'find', '\u0201')
@@ -288,7 +315,7 @@ class UnicodeTest(string_tests.CommonTest,
         self.checkequal(-1, '\u0102' * 100, 'find', '\u0102\U00100304')
 
     def test_rfind(self):
-        string_tests.CommonTest.test_rfind(self)
+        string_tests.StringLikeTest.test_rfind(self)
         # test implementation details of the memrchr fast path
         self.checkequal(0, '\u0102' + 'a' * 100 , 'rfind', '\u0102')
         self.checkequal(-1, '\u0102' + 'a' * 100 , 'rfind', '\u0201')
@@ -329,7 +356,7 @@ class UnicodeTest(string_tests.CommonTest,
         self.checkequal(-1, '\u0102' * 100, 'rfind', '\U00100304\u0102')
 
     def test_index(self):
-        string_tests.CommonTest.test_index(self)
+        string_tests.StringLikeTest.test_index(self)
         self.checkequalnofix(0, 'abcdefghiabc', 'index',  '')
         self.checkequalnofix(3, 'abcdefghiabc', 'index',  'def')
         self.checkequalnofix(0, 'abcdefghiabc', 'index',  'abc')
@@ -353,7 +380,7 @@ class UnicodeTest(string_tests.CommonTest,
         self.assertRaises(ValueError, ('\u0102' * 100).index, '\u0102\U00100304')
 
     def test_rindex(self):
-        string_tests.CommonTest.test_rindex(self)
+        string_tests.StringLikeTest.test_rindex(self)
         self.checkequalnofix(12, 'abcdefghiabc', 'rindex',  '')
         self.checkequalnofix(3,  'abcdefghiabc', 'rindex',  'def')
         self.checkequalnofix(9,  'abcdefghiabc', 'rindex',  'abc')
@@ -449,7 +476,7 @@ class UnicodeTest(string_tests.CommonTest,
         self.assertRaises(TypeError, 'abababc'.translate, 'abc', 'xyz')
 
     def test_split(self):
-        string_tests.CommonTest.test_split(self)
+        string_tests.StringLikeTest.test_split(self)
 
         # test mixed kinds
         for left, right in ('ba', '\u0101\u0100', '\U00010301\U00010300'):
@@ -466,7 +493,7 @@ class UnicodeTest(string_tests.CommonTest,
                                 left + delim * 2 + right, 'split', delim *2)
 
     def test_rsplit(self):
-        string_tests.CommonTest.test_rsplit(self)
+        string_tests.StringLikeTest.test_rsplit(self)
         # test mixed kinds
         for left, right in ('ba', 'юё', '\u0101\u0100', '\U00010301\U00010300'):
             left *= 9
@@ -486,7 +513,7 @@ class UnicodeTest(string_tests.CommonTest,
                              left + right, 'rsplit', None)
 
     def test_partition(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_partition(self)
+        string_tests.StringLikeTest.test_partition(self)
         # test mixed kinds
         self.checkequal(('ABCDEFGH', '', ''), 'ABCDEFGH', 'partition', '\u4200')
         for left, right in ('ba', '\u0101\u0100', '\U00010301\U00010300'):
@@ -503,7 +530,7 @@ class UnicodeTest(string_tests.CommonTest,
                                 left + delim * 2 + right, 'partition', delim * 2)
 
     def test_rpartition(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_rpartition(self)
+        string_tests.StringLikeTest.test_rpartition(self)
         # test mixed kinds
         self.checkequal(('', '', 'ABCDEFGH'), 'ABCDEFGH', 'rpartition', '\u4200')
         for left, right in ('ba', '\u0101\u0100', '\U00010301\U00010300'):
@@ -520,7 +547,7 @@ class UnicodeTest(string_tests.CommonTest,
                                 left + delim * 2 + right, 'rpartition', delim * 2)
 
     def test_join(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_join(self)
+        string_tests.StringLikeTest.test_join(self)
 
         class MyWrapper:
             def __init__(self, sval): self.sval = sval
@@ -548,7 +575,7 @@ class UnicodeTest(string_tests.CommonTest,
         self.assertRaises(OverflowError, ''.join, seq)
 
     def test_replace(self):
-        string_tests.CommonTest.test_replace(self)
+        string_tests.StringLikeTest.test_replace(self)
 
         # method call forwarded from str implementation because of unicode argument
         self.checkequalnofix('one@two!three!', 'one!two!three!', 'replace', '!', '@', 1)
@@ -831,6 +858,15 @@ class UnicodeTest(string_tests.CommonTest,
         self.assertTrue('\U0001F46F'.isprintable())
         self.assertFalse('\U000E0020'.isprintable())
 
+    @support.requires_resource('cpu')
+    def test_isprintable_invariant(self):
+        for codepoint in range(sys.maxunicode + 1):
+            char = chr(codepoint)
+            category = unicodedata.category(char)
+            self.assertEqual(char.isprintable(),
+                             category[0] not in ('C', 'Z')
+                             or char == ' ')
+
     def test_surrogates(self):
         for s in ('a\uD800b\uDFFF', 'a\uDFFFb\uD800',
                   'a\uD800b\uDFFFa', 'a\uDFFFb\uD800a'):
@@ -859,7 +895,7 @@ class UnicodeTest(string_tests.CommonTest,
 
 
     def test_lower(self):
-        string_tests.CommonTest.test_lower(self)
+        string_tests.StringLikeTest.test_lower(self)
         self.assertEqual('\U00010427'.lower(), '\U0001044F')
         self.assertEqual('\U00010427\U00010427'.lower(),
                          '\U0001044F\U0001044F')
@@ -890,7 +926,7 @@ class UnicodeTest(string_tests.CommonTest,
         self.assertEqual('\u00b5'.casefold(), '\u03bc')
 
     def test_upper(self):
-        string_tests.CommonTest.test_upper(self)
+        string_tests.StringLikeTest.test_upper(self)
         self.assertEqual('\U0001044F'.upper(), '\U00010427')
         self.assertEqual('\U0001044F\U0001044F'.upper(),
                          '\U00010427\U00010427')
@@ -909,7 +945,7 @@ class UnicodeTest(string_tests.CommonTest,
     # TODO: RUSTPYTHON
     @unittest.expectedFailure
     def test_capitalize(self):
-        string_tests.CommonTest.test_capitalize(self)
+        string_tests.StringLikeTest.test_capitalize(self)
         self.assertEqual('\U0001044F'.capitalize(), '\U00010427')
         self.assertEqual('\U0001044F\U0001044F'.capitalize(),
                          '\U00010427\U0001044F')
@@ -947,7 +983,7 @@ class UnicodeTest(string_tests.CommonTest,
     # TODO: RUSTPYTHON
     @unittest.expectedFailure
     def test_swapcase(self):
-        string_tests.CommonTest.test_swapcase(self)
+        string_tests.StringLikeTest.test_swapcase(self)
         self.assertEqual('\U0001044F'.swapcase(), '\U00010427')
         self.assertEqual('\U00010427'.swapcase(), '\U0001044F')
         self.assertEqual('\U0001044F\U0001044F'.swapcase(),
@@ -973,7 +1009,7 @@ class UnicodeTest(string_tests.CommonTest,
         self.assertEqual('\u1fd2'.swapcase(), '\u0399\u0308\u0300')
 
     def test_center(self):
-        string_tests.CommonTest.test_center(self)
+        string_tests.StringLikeTest.test_center(self)
         self.assertEqual('x'.center(2, '\U0010FFFF'),
                          'x\U0010FFFF')
         self.assertEqual('x'.center(3, '\U0010FFFF'),
@@ -1483,7 +1519,7 @@ class UnicodeTest(string_tests.CommonTest,
     # TODO: RUSTPYTHON
     @unittest.expectedFailure
     def test_formatting(self):
-        string_tests.MixinStrUnicodeUserStringTest.test_formatting(self)
+        string_tests.StringLikeTest.test_formatting(self)
         # Testing Unicode formatting strings...
         self.assertEqual("%s, %s" % ("abc", "abc"), 'abc, abc')
         self.assertEqual("%s, %s, %i, %f, %5.2f" % ("abc", "abc", 1, 2, 3), 'abc, abc, 1, 2.000000,  3.00')
@@ -1659,7 +1695,7 @@ class UnicodeTest(string_tests.CommonTest,
             self.assertIn('str', exc)
             self.assertIn('tuple', exc)
 
-    @support.run_with_locale('LC_ALL', 'de_DE', 'fr_FR')
+    @support.run_with_locale('LC_ALL', 'de_DE', 'fr_FR', '')
     def test_format_float(self):
         # should not format with a comma, but always with C locale
         self.assertEqual('1.0', '%.1f' % 1.0)
@@ -1730,8 +1766,6 @@ class UnicodeTest(string_tests.CommonTest,
             'character buffers are decoded to unicode'
         )
 
-        self.assertRaises(TypeError, str, 42, 42, 42)
-
     # TODO: RUSTPYTHON
     @unittest.expectedFailure
     def test_constructor_keyword_args(self):
@@ -1910,6 +1944,12 @@ class UnicodeTest(string_tests.CommonTest,
             self.assertRaises(UnicodeDecodeError,
                               (b'\xF4'+cb+b'\xBF\xBF').decode, 'utf-8')
 
+    def test_issue127903(self):
+        # gh-127903: ``_copy_characters`` crashes on DEBUG builds when
+        # there is nothing to copy.
+        d = datetime.datetime(2013, 11, 10, 14, 20, 59)
+        self.assertEqual(d.strftime('%z'), '')
+
     def test_issue8271(self):
         # Issue #8271: during the decoding of an invalid UTF-8 byte sequence,
         # only the start byte and the continuation byte(s) are now considered
@@ -2396,28 +2436,37 @@ class UnicodeTest(string_tests.CommonTest,
     @unittest.expectedFailure
     def test_conversion(self):
         # Make sure __str__() works properly
-        class ObjectToStr:
-            def __str__(self):
-                return "foo"
-
-        class StrSubclassToStr(str):
-            def __str__(self):
-                return "foo"
-
-        class StrSubclassToStrSubclass(str):
-            def __new__(cls, content=""):
-                return str.__new__(cls, 2*content)
-            def __str__(self):
+        class StrWithStr(str):
+            def __new__(cls, value):
+                self = str.__new__(cls, "")
+                self.value = value
                 return self
+            def __str__(self):
+                return self.value
 
-        self.assertEqual(str(ObjectToStr()), "foo")
-        self.assertEqual(str(StrSubclassToStr("bar")), "foo")
-        s = str(StrSubclassToStrSubclass("foo"))
-        self.assertEqual(s, "foofoo")
-        self.assertIs(type(s), StrSubclassToStrSubclass)
-        s = StrSubclass(StrSubclassToStrSubclass("foo"))
-        self.assertEqual(s, "foofoo")
-        self.assertIs(type(s), StrSubclass)
+        self.assertTypedEqual(str(WithStr('abc')), 'abc')
+        self.assertTypedEqual(str(WithStr(StrSubclass('abc'))), StrSubclass('abc'))
+        self.assertTypedEqual(StrSubclass(WithStr('abc')), StrSubclass('abc'))
+        self.assertTypedEqual(StrSubclass(WithStr(StrSubclass('abc'))),
+                              StrSubclass('abc'))
+        self.assertTypedEqual(StrSubclass(WithStr(OtherStrSubclass('abc'))),
+                              StrSubclass('abc'))
+
+        self.assertTypedEqual(str(StrWithStr('abc')), 'abc')
+        self.assertTypedEqual(str(StrWithStr(StrSubclass('abc'))), StrSubclass('abc'))
+        self.assertTypedEqual(StrSubclass(StrWithStr('abc')), StrSubclass('abc'))
+        self.assertTypedEqual(StrSubclass(StrWithStr(StrSubclass('abc'))),
+                              StrSubclass('abc'))
+        self.assertTypedEqual(StrSubclass(StrWithStr(OtherStrSubclass('abc'))),
+                              StrSubclass('abc'))
+
+        self.assertTypedEqual(str(WithRepr('<abc>')), '<abc>')
+        self.assertTypedEqual(str(WithRepr(StrSubclass('<abc>'))), StrSubclass('<abc>'))
+        self.assertTypedEqual(StrSubclass(WithRepr('<abc>')), StrSubclass('<abc>'))
+        self.assertTypedEqual(StrSubclass(WithRepr(StrSubclass('<abc>'))),
+                              StrSubclass('<abc>'))
+        self.assertTypedEqual(StrSubclass(WithRepr(OtherStrSubclass('<abc>'))),
+                              StrSubclass('<abc>'))
 
     def test_unicode_repr(self):
         class s1:
@@ -2652,6 +2701,49 @@ class UnicodeTest(string_tests.CommonTest,
         proc = assert_python_failure('-X', 'dev', '-c', code)
         self.assertEqual(proc.rc, 10, proc)
 
+    # TODO: RUSTPYTHON
+    @unittest.expectedFailure
+    def test_str_invalid_call(self):
+        # too many args
+        with self.assertRaisesRegex(TypeError, r"str expected at most 3 arguments, got 4"):
+            str("too", "many", "argu", "ments")
+        with self.assertRaisesRegex(TypeError, r"str expected at most 3 arguments, got 4"):
+            str(1, "", "", 1)
+
+        # no such kw arg
+        with self.assertRaisesRegex(TypeError, r"str\(\) got an unexpected keyword argument 'test'"):
+            str(test=1)
+
+        # 'encoding' must be str
+        with self.assertRaisesRegex(TypeError, r"str\(\) argument 'encoding' must be str, not int"):
+            str(1, 1)
+        with self.assertRaisesRegex(TypeError, r"str\(\) argument 'encoding' must be str, not int"):
+            str(1, encoding=1)
+        with self.assertRaisesRegex(TypeError, r"str\(\) argument 'encoding' must be str, not bytes"):
+            str(b"x", b"ascii")
+        with self.assertRaisesRegex(TypeError, r"str\(\) argument 'encoding' must be str, not bytes"):
+            str(b"x", encoding=b"ascii")
+
+        # 'errors' must be str
+        with self.assertRaisesRegex(TypeError, r"str\(\) argument 'encoding' must be str, not int"):
+            str(1, 1, 1)
+        with self.assertRaisesRegex(TypeError, r"str\(\) argument 'errors' must be str, not int"):
+            str(1, errors=1)
+        with self.assertRaisesRegex(TypeError, r"str\(\) argument 'errors' must be str, not int"):
+            str(1, "", errors=1)
+        with self.assertRaisesRegex(TypeError, r"str\(\) argument 'errors' must be str, not bytes"):
+            str(b"x", "ascii", b"strict")
+        with self.assertRaisesRegex(TypeError, r"str\(\) argument 'errors' must be str, not bytes"):
+            str(b"x", "ascii", errors=b"strict")
+
+        # both positional and kwarg
+        with self.assertRaisesRegex(TypeError, r"argument for str\(\) given by name \('encoding'\) and position \(2\)"):
+            str(b"x", "utf-8", encoding="ascii")
+        with self.assertRaisesRegex(TypeError, r"str\(\) takes at most 3 arguments \(4 given\)"):
+            str(b"x", "utf-8", "ignore", encoding="ascii")
+        with self.assertRaisesRegex(TypeError, r"str\(\) takes at most 3 arguments \(4 given\)"):
+            str(b"x", "utf-8", "strict", errors="ignore")
+
 
 class StringModuleTest(unittest.TestCase):
     def test_formatter_parser(self):

Lib/test/test_unicode_file.py

@@ -110,7 +110,7 @@ class TestUnicodeFiles(unittest.TestCase):
             os.unlink(filename)
         self.assertTrue(not os.path.exists(filename))
         # and again with os.open.
-        f = os.open(filename, os.O_CREAT)
+        f = os.open(filename, os.O_CREAT | os.O_WRONLY)
         os.close(f)
         try:
             self._do_single(filename)

Lib/test/test_unicode_file_functions.py

@@ -5,7 +5,7 @@ import sys
 import unittest
 import warnings
 from unicodedata import normalize
-from test.support import os_helper
+from test.support import is_apple, os_helper
 from test import support
 
 
@@ -23,13 +23,13 @@ filenames = [
     '10_\u1fee\u1ffd',
     ]
 
-# Mac OS X decomposes Unicode names, using Normal Form D.
+# Apple platforms decompose Unicode names, using Normal Form D.
 # http://developer.apple.com/mac/library/qa/qa2001/qa1173.html
 # "However, most volume formats do not follow the exact specification for
 # these normal forms.  For example, HFS Plus uses a variant of Normal Form D
 # in which U+2000 through U+2FFF, U+F900 through U+FAFF, and U+2F800 through
 # U+2FAFF are not decomposed."
-if sys.platform != 'darwin':
+if not is_apple:
     filenames.extend([
         # Specific code points: NFC(fn), NFD(fn), NFKC(fn) and NFKD(fn) all different
         '11_\u0385\u03d3\u03d4',
@@ -119,11 +119,11 @@ class UnicodeFileTests(unittest.TestCase):
             os.stat(name)
             self._apply_failure(os.listdir, name, self._listdir_failure)
 
-    # Skip the test on darwin, because darwin does normalize the filename to
+    # Skip the test on Apple platforms, because they don't normalize the filename to
     # NFD (a variant of Unicode NFD form). Normalize the filename to NFC, NFKC,
     # NFKD in Python is useless, because darwin will normalize it later and so
     # open(), os.stat(), etc. don't raise any exception.
-    @unittest.skipIf(sys.platform == 'darwin', 'irrelevant test on Mac OS X')
+    @unittest.skipIf(is_apple, 'irrelevant test on Apple platforms')
     @unittest.skipIf(
         support.is_emscripten or support.is_wasi,
         "test fails on Emscripten/WASI when host platform is macOS."
@@ -142,10 +142,10 @@ class UnicodeFileTests(unittest.TestCase):
             self._apply_failure(os.remove, name)
             self._apply_failure(os.listdir, name)
 
-    # Skip the test on darwin, because darwin uses a normalization different
+    # Skip the test on Apple platforms, because they use a normalization different
     # than Python NFD normalization: filenames are different even if we use
     # Python NFD normalization.
-    @unittest.skipIf(sys.platform == 'darwin', 'irrelevant test on Mac OS X')
+    @unittest.skipIf(is_apple, 'irrelevant test on Apple platforms')
     def test_listdir(self):
         sf0 = set(self.files)
         with warnings.catch_warnings():

Lib/test/test_unicode_identifiers.py

@@ -21,7 +21,7 @@ class PEP3131Test(unittest.TestCase):
     @unittest.expectedFailure
     def test_invalid(self):
         try:
-            from test import badsyntax_3131
+            from test.tokenizedata import badsyntax_3131
         except SyntaxError as err:
             self.assertEqual(str(err),
               "invalid character '€' (U+20AC) (badsyntax_3131.py, line 2)")

Lib/test/test_unicodedata.py

@@ -11,15 +11,20 @@ from http.client import HTTPException
 import sys
 import unicodedata
 import unittest
-from test.support import (open_urlresource, requires_resource, script_helper,
-                          cpython_only, check_disallow_instantiation,
-                          ResourceDenied)
+from test.support import (
+    open_urlresource,
+    requires_resource,
+    script_helper,
+    cpython_only,
+    check_disallow_instantiation,
+    force_not_colorized,
+)
 
 
 class UnicodeMethodsTest(unittest.TestCase):
 
     # update this, if the database changes
-    expectedchecksum = '4739770dd4d0e5f1b1677accfc3552ed3c8ef326'
+    expectedchecksum = '63aa77dcb36b0e1df082ee2a6071caeda7f0955e'
 
     # TODO: RUSTPYTHON
     @unittest.expectedFailure
@@ -74,7 +79,8 @@ class UnicodeFunctionsTest(UnicodeDatabaseTest):
 
     # Update this if the database changes. Make sure to do a full rebuild
     # (e.g. 'make distclean && make') to get the correct checksum.
-    expectedchecksum = '98d602e1f69d5c5bb8a5910c40bbbad4e18e8370'
+    expectedchecksum = '232affd2a50ec4bd69d2482aa0291385cbdefaba'
+
     # TODO: RUSTPYTHON
     @unittest.expectedFailure
     @requires_resource('cpu')
@@ -94,6 +100,8 @@ class UnicodeFunctionsTest(UnicodeDatabaseTest):
                 self.db.decomposition(char),
                 str(self.db.mirrored(char)),
                 str(self.db.combining(char)),
+                unicodedata.east_asian_width(char),
+                self.db.name(char, ""),
             ]
             h.update(''.join(data).encode("ascii"))
         result = h.hexdigest()
@@ -106,6 +114,28 @@ class UnicodeFunctionsTest(UnicodeDatabaseTest):
             if looked_name := self.db.name(char, None):
                 self.assertEqual(self.db.lookup(looked_name), char)
 
+    def test_no_names_in_pua(self):
+        puas = [*range(0xe000, 0xf8ff),
+                *range(0xf0000, 0xfffff),
+                *range(0x100000, 0x10ffff)]
+        for i in puas:
+            char = chr(i)
+            self.assertRaises(ValueError, self.db.name, char)
+
+    # TODO: RUSTPYTHON; LookupError: undefined character name 'LATIN SMLL LETR A'
+    @unittest.expectedFailure
+    def test_lookup_nonexistant(self):
+        # just make sure that lookup can fail
+        for nonexistant in [
+            "LATIN SMLL LETR A",
+            "OPEN HANDS SIGHS",
+            "DREGS",
+            "HANDBUG",
+            "MODIFIER LETTER CYRILLIC SMALL QUESTION MARK",
+            "???",
+        ]:
+            self.assertRaises(KeyError, self.db.lookup, nonexistant)
+
     # TODO: RUSTPYTHON
     @unittest.expectedFailure
     def test_digit(self):
@@ -245,6 +275,25 @@ class UnicodeFunctionsTest(UnicodeDatabaseTest):
         self.assertEqual(eaw('\u2010'), 'A')
         self.assertEqual(eaw('\U00020000'), 'W')
 
+    # TODO: RUSTPYTHON
+    @unittest.expectedFailure
+    def test_east_asian_width_unassigned(self):
+        eaw = self.db.east_asian_width
+        # unassigned
+        for char in '\u0530\u0ecf\u10c6\u20fc\uaaca\U000107bd\U000115f2':
+            self.assertEqual(eaw(char), 'N')
+            self.assertIs(self.db.name(char, None), None)
+
+        # unassigned but reserved for CJK
+        for char in '\uFA6E\uFADA\U0002A6E0\U0002FA20\U0003134B\U0003FFFD':
+            self.assertEqual(eaw(char), 'W')
+            self.assertIs(self.db.name(char, None), None)
+
+        # private use areas
+        for char in '\uE000\uF800\U000F0000\U000FFFEE\U00100000\U0010FFF0':
+            self.assertEqual(eaw(char), 'A')
+            self.assertIs(self.db.name(char, None), None)
+
     # TODO: RUSTPYTHON
     @unittest.expectedFailure
     def test_east_asian_width_9_0_changes(self):
@@ -260,6 +309,7 @@ class UnicodeMiscTest(UnicodeDatabaseTest):
 
     # TODO: RUSTPYTHON
     @unittest.expectedFailure
+    @force_not_colorized
     def test_failed_import_during_compiling(self):
         # Issue 4367
         # Decoding \N escapes requires the unicodedata module. If it can't be
@@ -322,6 +372,7 @@ class UnicodeMiscTest(UnicodeDatabaseTest):
         self.assertTrue("\u1d79".upper()=='\ua77d')
         self.assertTrue(".".upper()=='.')
 
+    @requires_resource('cpu')
     def test_bug_5828(self):
         self.assertEqual("\u1d79".lower(), "\u1d79")
         # Only U+0000 should have U+0000 as its upper/lower/titlecase variant
@@ -364,6 +415,7 @@ class NormalizationTest(unittest.TestCase):
         return "".join([chr(x) for x in data])
 
     @requires_resource('network')
+    @requires_resource('cpu')
     def test_normalization(self):
         TESTDATAFILE = "NormalizationTest.txt"
         TESTDATAURL = f"http://www.pythontest.net/unicode/{unicodedata.unidata_version}/{TESTDATAFILE}"

Lib/test/test_userstring.py

@@ -7,8 +7,7 @@ from test import string_tests
 from collections import UserString
 
 class UserStringTest(
-    string_tests.CommonTest,
-    string_tests.MixinStrUnicodeUserStringTest,
+    string_tests.StringLikeTest,
     unittest.TestCase
     ):