[PavlNekrasov]

1. Lib/idlelib/redirector.py — repr shows a spurious tuple

The comma inside the substitution field makes the expression a one-element
tuple, so repr() returns ('WidgetRedirector',)(Text<.text>) instead of
WidgetRedirector(Text<.text>). The same mistake is duplicated in
OriginalCommand.__repr__ in the same file.

cpython/Lib/idlelib/redirector.py

Lines 49 to 51 in 288cc2b

	def __repr__(self):
	w = self.widget
	return f"{self.__class__.__name__,}({w.__class__.__name__}<{w._w}>)"

cpython/Lib/idlelib/redirector.py

Lines 145 to 146 in 288cc2b

	def __repr__(self):
	return f"{self.__class__.__name__,}({self.redir!r}, {self.operation!r})"

2. Lib/test/test_email/test_message.py — header never appended

cpython/Lib/test/test_email/test_message.py

Lines 584 to 585 in 288cc2b

	if subtype != 'no_content':
	('content-shadow', 'Logrus'),

The tuple is created and immediately discarded, so content-shadow never ends
up in msg_headers and the check that non-standard content-* headers (other
than Content-Type) are copied to the part in _check_make_multipart never
runs:

cpython/Lib/test/test_email/test_message.py

Lines 616 to 619 in 288cc2b

	for name, value in msg_headers:
	if name.lower().startswith('content-'):
	self.assertEqual(part[name], value)
	count += 1

The mistake masks a real regression: if the condition in Lib/email/message.py
is narrowed to Content-Type only, the current test set stays green (307 OK),
while with the content-shadow header properly appended 36 tests fail.

Found by Linux Verification Center (linuxtesting.org) with SVACE

[bedevere-app]

Most changes to Python require a NEWS entry. Add one using the blurb_it web app or the blurb command-line tool.

If this change has little impact on Python users, wait for a maintainer to apply the skip news label instead.

[picnixz]

Please create an issue.

[picnixz]

These are also two unrelated changes so you need two distinct PR.

[terryjreedy]

Yes, 2 PRs. I will handle the IDLE part and someone else the email issue.

[terryjreedy]

@PavlNekrasov I figured out why the IDLE test passed without and with the error. The bad repr method does not affect the purpose of the class, but I will merge the proposed change when separately available.

This should actually be 2 issues. While it is true that the stray comma in the email test wraps the 2-tuple in a 1-tuple, the difference in the expression statement is not relevant as the result is not bound and disappears. The important proposed change is appending the 2-tuple to the message. If the test passes without (now) or with (proposed), then something seems amiss. At least in the test. An email person should decide.