[3.12] gh-117313: Fix re-folding email messages containing non-standard line separators (GH-117369) (GH-117971)

Only treat '\n', '\r' and '\r\n' as line separators in re-folding the email
messages.  Preserve control characters '\v', '\f', '\x1c', '\x1d' and '\x1e'
and Unicode line separators '\x85', '\u2028' and '\u2029' as is.
(cherry picked from commit aec1dac4ef)

Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>

Lib/email/policy.py

@@ -21,7 +21,7 @@ __all__ = [
     'HTTP',
     ]
 
-linesep_splitter = re.compile(r'\n|\r')
+linesep_splitter = re.compile(r'\n|\r\n?')
 
 @_extend_docstrings
 class EmailPolicy(Policy):
@@ -205,7 +205,8 @@ class EmailPolicy(Policy):
         if hasattr(value, 'name'):
             return value.fold(policy=self)
         maxlen = self.max_line_length if self.max_line_length else sys.maxsize
-        lines = value.splitlines()
+        # We can't use splitlines here because it splits on more than \r and \n.
+        lines = linesep_splitter.split(value)
         refold = (self.refold_source == 'all' or
                   self.refold_source == 'long' and
                     (lines and len(lines[0])+len(name)+2 > maxlen or

Lib/test/test_email/test_generator.py

@@ -140,6 +140,39 @@ class TestGeneratorBase:
         g.flatten(msg, linesep='\n')
         self.assertEqual(s.getvalue(), self.typ(expected))
 
+    def test_flatten_linesep(self):
+        source = 'Subject: one\n two\r three\r\n four\r\n\r\ntest body\r\n'
+        msg = self.msgmaker(self.typ(source))
+        self.assertEqual(msg['Subject'], 'one two three four')
+
+        expected = 'Subject: one\n two\n three\n four\n\ntest body\n'
+        s = self.ioclass()
+        g = self.genclass(s)
+        g.flatten(msg)
+        self.assertEqual(s.getvalue(), self.typ(expected))
+
+        expected = 'Subject: one two three four\n\ntest body\n'
+        s = self.ioclass()
+        g = self.genclass(s, policy=self.policy.clone(refold_source='all'))
+        g.flatten(msg)
+        self.assertEqual(s.getvalue(), self.typ(expected))
+
+    def test_flatten_control_linesep(self):
+        source = 'Subject: one\v two\f three\x1c four\x1d five\x1e six\r\n\r\ntest body\r\n'
+        msg = self.msgmaker(self.typ(source))
+        self.assertEqual(msg['Subject'], 'one\v two\f three\x1c four\x1d five\x1e six')
+
+        expected = 'Subject: one\v two\f three\x1c four\x1d five\x1e six\n\ntest body\n'
+        s = self.ioclass()
+        g = self.genclass(s)
+        g.flatten(msg)
+        self.assertEqual(s.getvalue(), self.typ(expected))
+
+        s = self.ioclass()
+        g = self.genclass(s, policy=self.policy.clone(refold_source='all'))
+        g.flatten(msg)
+        self.assertEqual(s.getvalue(), self.typ(expected))
+
     def test_set_mangle_from_via_policy(self):
         source = textwrap.dedent("""\
             Subject: test that
@@ -224,6 +257,22 @@ class TestGenerator(TestGeneratorBase, TestEmailBase):
     ioclass = io.StringIO
     typ = str
 
+    def test_flatten_unicode_linesep(self):
+        source = 'Subject: one\x85 two\u2028 three\u2029 four\r\n\r\ntest body\r\n'
+        msg = self.msgmaker(self.typ(source))
+        self.assertEqual(msg['Subject'], 'one\x85 two\u2028 three\u2029 four')
+
+        expected = 'Subject: =?utf-8?b?b25lwoUgdHdv4oCoIHRocmVl4oCp?= four\n\ntest body\n'
+        s = self.ioclass()
+        g = self.genclass(s)
+        g.flatten(msg)
+        self.assertEqual(s.getvalue(), self.typ(expected))
+
+        s = self.ioclass()
+        g = self.genclass(s, policy=self.policy.clone(refold_source='all'))
+        g.flatten(msg)
+        self.assertEqual(s.getvalue(), self.typ(expected))
+
 
 class TestBytesGenerator(TestGeneratorBase, TestEmailBase):
 

Misc/NEWS.d/next/Library/2024-03-29-15-14-51.gh-issue-117313.ks_ONu.rst

@@ -0,0 +1,4 @@
+Only treat ``'\n'``, ``'\r'`` and ``'\r\n'`` as line separators in
+re-folding the :mod:`email` messages. Preserve control characters ``'\v'``,
+``'\f'``, ``'\x1c'``, ``'\x1d'`` and ``'\x1e'`` and Unicode line separators
+``'\x85'``, ``'\u2028'`` and ``'\u2029'`` as is.