gh-105285: Use UTF-8 on email content when CTE is '8bit'

Lib/email/message.py

@@ -302,7 +302,10 @@ def get_payload(self, i=None, decode=False):
             return payload
         if isinstance(payload, str):
             try:
-                bpayload = payload.encode('ascii', 'surrogateescape')
+                if cte == '8bit':
+                    bpayload = payload.encode(self.get_param('charset', 'ascii'), 'surrogateescape')
+                else:
+                    bpayload = payload.encode('ascii', 'surrogateescape')
             except UnicodeEncodeError:
                 # This won't happen for RFC compliant messages (messages
                 # containing only ASCII code points in the unicode input).

Lib/test/test_email/test_email.py

@@ -673,7 +673,7 @@ def test_broken_unicode_payload(self):
         x = 'this is a br\xf6ken thing to do'
         msg = Message()
         msg['content-type'] = 'text/plain'
-        msg['content-transfer-encoding'] = '8bit'
+        msg['content-transfer-encoding'] = '7bit'
         msg.set_payload(x)
         self.assertEqual(msg.get_payload(decode=True),
                          bytes(x, 'raw-unicode-escape'))
@@ -750,6 +750,15 @@ def test_binary_base64_payload(self):
                 b'foo\xe6\x96\x87bar',
                 'get_payload returns wrong result with charset %s.' % charset)
 
+    def test_8bit_utf8_payload(self):
+        x = 'this is the hötel'
+        msg = Message()
+        msg['content-type'] = 'text/plain'
+        msg['content-transfer-encoding'] = '8bit'
+        msg.set_payload(x)
+        self.assertEqual(msg.get_payload(decode=True),
+                         bytes(x, 'utf-8'))
+
     def test_binary_uuencode_payload(self):
         for charset in ('latin-1', 'ascii'):
             for encoding in ('x-uuencode', 'uuencode', 'uue', 'x-uue'):

Misc/NEWS.d/next/Library/2023-06-05-02-37-52.gh-issue-105285.fwUxuR.rst

@@ -0,0 +1 @@
+Attempt to decode emails with UTF-8 when the CTE is ``8bit``.