Merge branch 'main' into narrowing-to-constants-in-branches-with-is

Author: reidenong

Doc/library/stdtypes.rst

@@ -2562,6 +2562,19 @@ expression support in the :mod:`re` module).
    after the separator.  If the separator is not found, return a 3-tuple containing
    two empty strings, followed by the string itself.
 
+   For example:
+
+   .. doctest::
+
+      >>> 'Monty Python'.rpartition(' ')
+      ('Monty', ' ', 'Python')
+      >>> "Monty Python's Flying Circus".rpartition(' ')
+      ("Monty Python's Flying", ' ', 'Circus')
+      >>> 'Monty Python'.rpartition('-')
+      ('', '', 'Monty Python')
+
+   See also :meth:`partition`.
+
 
 .. method:: str.rsplit(sep=None, maxsplit=-1)
 

Doc/library/tkinter.rst

@@ -177,12 +177,12 @@ the modern themed widget set and API::
    .. attribute:: master
 
       The widget object that contains this widget.  For :class:`Tk`, the
-      *master* is :const:`None` because it is the main window.  The terms
+      :attr:`!master` is :const:`None` because it is the main window.  The terms
       *master* and *parent* are similar and sometimes used interchangeably
       as argument names; however, calling :meth:`winfo_parent` returns a
-      string of the widget name whereas :attr:`master` returns the object.
+      string of the widget name whereas :attr:`!master` returns the object.
       *parent*/*child* reflects the tree-like relationship while
-      *master*/*slave* reflects the container structure.
+      *master* (or *container*)/*content* reflects the container structure.
 
    .. attribute:: children
 
@@ -638,15 +638,15 @@ The Packer
 .. index:: single: packing (widgets)
 
 The packer is one of Tk's geometry-management mechanisms.    Geometry managers
-are used to specify the relative positioning of widgets within their container -
-their mutual *master*.  In contrast to the more cumbersome *placer* (which is
+are used to specify the relative positioning of widgets within their container.
+In contrast to the more cumbersome *placer* (which is
 used less commonly, and we do not cover here), the packer takes qualitative
 relationship specification - *above*, *to the left of*, *filling*, etc - and
 works everything out to determine the exact placement coordinates for you.
 
-The size of any *master* widget is determined by the size of the "slave widgets"
-inside.  The packer is used to control where slave widgets appear inside the
-master into which they are packed.  You can pack widgets into frames, and frames
+The size of any container widget is determined by the size of the "content widgets"
+inside.  The packer is used to control where content widgets appear inside the
+container into which they are packed.  You can pack widgets into frames, and frames
 into other frames, in order to achieve the kind of layout you desire.
 Additionally, the arrangement is dynamically adjusted to accommodate incremental
 changes to the configuration, once it is packed.
@@ -673,7 +673,7 @@ For more extensive information on the packer and the options that it can take,
 see the man pages and page 183 of John Ousterhout's book.
 
 anchor
-   Anchor type.  Denotes where the packer is to place each slave in its parcel.
+   Anchor type.  Denotes where the packer is to place each content in its parcel.
 
 expand
    Boolean, ``0`` or ``1``.
@@ -682,10 +682,10 @@ fill
    Legal values: ``'x'``, ``'y'``, ``'both'``, ``'none'``.
 
 ipadx and ipady
-   A distance - designating internal padding on each side of the slave widget.
+   A distance - designating internal padding on each side of the content.
 
 padx and pady
-   A distance - designating external padding on each side of the slave widget.
+   A distance - designating external padding on each side of the content.
 
 side
    Legal values are: ``'left'``, ``'right'``, ``'top'``, ``'bottom'``.
@@ -758,8 +758,8 @@ subclassed from the :class:`Wm` class, and so can call the :class:`Wm` methods
 directly.
 
 To get at the toplevel window that contains a given widget, you can often just
-refer to the widget's master.  Of course if the widget has been packed inside of
-a frame, the master won't represent a toplevel window.  To get at the toplevel
+refer to the widget's :attr:`master`.  Of course if the widget has been packed inside of
+a frame, the :attr:`!master` won't represent a toplevel window.  To get at the toplevel
 window that contains an arbitrary widget, you can call the :meth:`_root` method.
 This method begins with an underscore to denote the fact that this function is
 part of the implementation, and not an interface to Tk functionality.

Lib/email/_header_value_parser.py

@@ -101,6 +101,12 @@ def make_quoted_pairs(value):
     return str(value).replace('\\', '\\\\').replace('"', '\\"')
 
 
+def make_parenthesis_pairs(value):
+    """Escape parenthesis and backslash for use within a comment."""
+    return str(value).replace('\\', '\\\\') \
+        .replace('(', '\\(').replace(')', '\\)')
+
+
 def quote_string(value):
     escaped = make_quoted_pairs(value)
     return f'"{escaped}"'
@@ -943,7 +949,7 @@ def value(self):
         return ' '
 
     def startswith_fws(self):
-        return True
+        return self and self[0] in WSP
 
 
 class ValueTerminal(Terminal):
@@ -2963,6 +2969,13 @@ def _refold_parse_tree(parse_tree, *, policy):
                     [ValueTerminal(make_quoted_pairs(p), 'ptext')
                      for p in newparts] +
                     [ValueTerminal('"', 'ptext')])
+            if part.token_type == 'comment':
+                newparts = (
+                    [ValueTerminal('(', 'ptext')] +
+                    [ValueTerminal(make_parenthesis_pairs(p), 'ptext')
+                     if p.token_type == 'ptext' else p
+                     for p in newparts] +
+                    [ValueTerminal(')', 'ptext')])
             if not part.as_ew_allowed:
                 wrap_as_ew_blocked += 1
                 newparts.append(end_ew_not_allowed)

Lib/test/test_email/test__header_value_parser.py

@@ -3294,6 +3294,29 @@ def test_address_list_with_specials_in_long_quoted_string(self):
             with self.subTest(to=to):
                 self._test(parser.get_address_list(to)[0], folded, policy=policy)
 
+    def test_address_list_with_long_unwrapable_comment(self):
+        policy = self.policy.clone(max_line_length=40)
+        cases = [
+            # (to, folded)
+            ('(loremipsumdolorsitametconsecteturadipi)<spy@example.org>',
+             '(loremipsumdolorsitametconsecteturadipi)<spy@example.org>\n'),
+            ('<spy@example.org>(loremipsumdolorsitametconsecteturadipi)',
+             '<spy@example.org>(loremipsumdolorsitametconsecteturadipi)\n'),
+            ('(loremipsum dolorsitametconsecteturadipi)<spy@example.org>',
+             '(loremipsum dolorsitametconsecteturadipi)<spy@example.org>\n'),
+             ('<spy@example.org>(loremipsum dolorsitametconsecteturadipi)',
+             '<spy@example.org>(loremipsum\n dolorsitametconsecteturadipi)\n'),
+            ('(Escaped \\( \\) chars \\\\ in comments stay escaped)<spy@example.org>',
+             '(Escaped \\( \\) chars \\\\ in comments stay\n escaped)<spy@example.org>\n'),
+            ('((loremipsum)(loremipsum)(loremipsum)(loremipsum))<spy@example.org>',
+             '((loremipsum)(loremipsum)(loremipsum)(loremipsum))<spy@example.org>\n'),
+            ('((loremipsum)(loremipsum)(loremipsum) (loremipsum))<spy@example.org>',
+             '((loremipsum)(loremipsum)(loremipsum)\n (loremipsum))<spy@example.org>\n'),
+        ]
+        for (to, folded) in cases:
+            with self.subTest(to=to):
+                self._test(parser.get_address_list(to)[0], folded, policy=policy)
+
     # XXX Need tests with comments on various sides of a unicode token,
     # and with unicode tokens in the comments.  Spaces inside the quotes
     # currently don't do the right thing.

Lib/test/test_io/test_bufferedio.py

@@ -10,7 +10,7 @@
 from collections import deque, UserList
 from itertools import cycle, count
 from test import support
-from test.support import os_helper, threading_helper
+from test.support import check_sanitizer, os_helper, threading_helper
 from .utils import byteslike, CTestCase, PyTestCase
 
 
@@ -623,6 +623,25 @@ def test_bad_readinto_type(self):
             bufio.readline()
         self.assertIsInstance(cm.exception.__cause__, TypeError)
 
+    @unittest.skipUnless(sys.maxsize > 2**32, 'requires 64bit platform')
+    @unittest.skipIf(check_sanitizer(thread=True),
+                     'ThreadSanitizer aborts on huge allocations (exit code 66).')
+    def test_read1_error_does_not_cause_reentrant_failure(self):
+        self.addCleanup(os_helper.unlink, os_helper.TESTFN)
+        with self.open(os_helper.TESTFN, "wb") as f:
+            f.write(b"hello")
+
+        with self.open(os_helper.TESTFN, "rb", buffering=0) as raw:
+            bufio = self.tp(raw, buffer_size=8)
+            # To request a size that is far too huge to ever be satisfied,
+            # so that the internal buffer allocation reliably fails with MemoryError.
+            huge = sys.maxsize // 2 + 1
+            with self.assertRaises(MemoryError):
+                bufio.read1(huge)
+
+            # Used to crash before gh-143689:
+            self.assertEqual(bufio.read1(1), b"h")
+
 
 class PyBufferedReaderTest(BufferedReaderTest, PyTestCase):
     tp = pyio.BufferedReader

Lib/test/test_pathlib/test_write.py

@@ -65,15 +65,17 @@ def test_open_wb(self):
 
     def test_write_bytes(self):
         p = self.root / 'fileA'
-        p.write_bytes(b'abcdefg')
-        self.assertEqual(self.ground.readbytes(p), b'abcdefg')
+        data = b'abcdefg'
+        self.assertEqual(len(data), p.write_bytes(data))
+        self.assertEqual(self.ground.readbytes(p), data)
         # Check that trying to write str does not truncate the file.
         self.assertRaises(TypeError, p.write_bytes, 'somestr')
-        self.assertEqual(self.ground.readbytes(p), b'abcdefg')
+        self.assertEqual(self.ground.readbytes(p), data)
 
     def test_write_text(self):
         p = self.root / 'fileA'
-        p.write_text('äbcdefg', encoding='latin-1')
+        data = 'äbcdefg'
+        self.assertEqual(len(data), p.write_text(data, encoding='latin-1'))
         self.assertEqual(self.ground.readbytes(p), b'\xe4bcdefg')
         # Check that trying to write bytes does not truncate the file.
         self.assertRaises(TypeError, p.write_text, b'somebytes', encoding='utf-8')

Lib/test/test_unicodedata.py

@@ -83,15 +83,7 @@ def test_method_checksum(self):
         self.assertEqual(result, self.expectedchecksum)
 
 
-class UnicodeFunctionsTest(unittest.TestCase):
-    db = unicodedata
-    old = False
-
-    # Update this if the database changes. Make sure to do a full rebuild
-    # (e.g. 'make distclean && make') to get the correct checksum.
-    expectedchecksum = ('83cc43a2fbb779185832b4c049217d80b05bf349'
-                        if quicktest else
-                        '65670ae03a324c5f9e826a4de3e25bae4d73c9b7')
+class BaseUnicodeFunctionsTest:
 
     def test_function_checksum(self):
         db = self.db
@@ -589,6 +581,16 @@ def test_east_asian_width_unassigned(self):
             self.assertEqual(eaw(char), 'A')
             self.assertIs(self.db.name(char, None), None)
 
+class UnicodeFunctionsTest(unittest.TestCase, BaseUnicodeFunctionsTest):
+    db = unicodedata
+    old = False
+
+    # Update this if the database changes. Make sure to do a full rebuild
+    # (e.g. 'make distclean && make') to get the correct checksum.
+    expectedchecksum = ('83cc43a2fbb779185832b4c049217d80b05bf349'
+                        if quicktest else
+                        '65670ae03a324c5f9e826a4de3e25bae4d73c9b7')
+
     def test_isxidstart(self):
         self.assertTrue(self.db.isxidstart('S'))
         self.assertTrue(self.db.isxidstart('\u0AD0'))  # GUJARATI OM
@@ -832,18 +834,13 @@ def graphemes(*args):
             ['a', '\U0001F1FA\U0001F1E6', '\U0001F1FA\U0001F1F3'])
 
 
-class Unicode_3_2_0_FunctionsTest(UnicodeFunctionsTest):
+class Unicode_3_2_0_FunctionsTest(unittest.TestCase, BaseUnicodeFunctionsTest):
     db = unicodedata.ucd_3_2_0
     old = True
     expectedchecksum = ('f4526159891a4b766dd48045646547178737ba09'
                         if quicktest else
                         'f217b8688d7bdff31db4207e078a96702f091597')
 
-    test_grapheme_cluster_break = None
-    test_indic_conjunct_break = None
-    test_extended_pictographic = None
-    test_grapheme_break = None
-
 
 class UnicodeMiscTest(unittest.TestCase):
     db = unicodedata