1berry/cpython
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

gh-76511: Fix email.Message.as_string() for non-ASCII message with ASCII charset (GH-116125)

Browse Source
This commit is contained in:
Serhiy Storchaka 2024-03-05 17:49:01 +02:00 committed by GitHub
parent df59401108
commit f97f25ef5d
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 21 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Lib/email/generator.py
View File

@ -243,7 +243,7 @@ class Generator:
# existing message. # existing message.
msg = deepcopy(msg) msg = deepcopy(msg)
del msg['content-transfer-encoding'] del msg['content-transfer-encoding']
msg.set_payload(payload, charset) msg.set_payload(msg._payload, charset)
payload = msg.get_payload() payload = msg.get_payload()
self._munge_cte = (msg['content-transfer-encoding'], self._munge_cte = (msg['content-transfer-encoding'],
msg['content-type']) msg['content-type'])

2
Lib/email/message.py
View File

@ -340,7 +340,7 @@ class Message:
return return
if not isinstance(charset, Charset): if not isinstance(charset, Charset):
charset = Charset(charset) charset = Charset(charset)
payload = payload.encode(charset.output_charset) payload = payload.encode(charset.output_charset, 'surrogateescape')
if hasattr(payload, 'decode'): if hasattr(payload, 'decode'):
self._payload = payload.decode('ascii', 'surrogateescape') self._payload = payload.decode('ascii', 'surrogateescape')
else: else:

15
Lib/test/test_email/test_email.py
View File

@ -337,6 +337,21 @@ class TestMessageAPI(TestEmailBase):
msg = email.message_from_bytes(source) msg = email.message_from_bytes(source)
self.assertEqual(msg.as_string(), expected) self.assertEqual(msg.as_string(), expected)
def test_nonascii_as_string_with_ascii_charset(self):
m = textwrap.dedent("""\
MIME-Version: 1.0
Content-type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 8bit
Test if non-ascii messages with no Content-Transfer-Encoding set
can be as_string'd:
Föö bär
""")
source = m.encode('iso-8859-1')
expected = source.decode('ascii', 'replace')
msg = email.message_from_bytes(source)
self.assertEqual(msg.as_string(), expected)
def test_nonascii_as_string_without_content_type_and_cte(self): def test_nonascii_as_string_without_content_type_and_cte(self):
m = textwrap.dedent("""\ m = textwrap.dedent("""\
MIME-Version: 1.0 MIME-Version: 1.0

4
Misc/NEWS.d/next/Library/2024-02-29-17-06-54.gh-issue-76511.WqjRLP.rst Normal file
View File

@ -0,0 +1,4 @@
Fix UnicodeEncodeError in :meth:`email.Message.as_string` that results when
a message that claims to be in the ascii character set actually has non-ascii
characters. Non-ascii characters are now replaced with the U+FFFD replacement
character, like in the ``replace`` error handler.