Merge #15249: Mangle From lines correctly when body contains invalid bytes.

bitdancer · bitdancer · commit ad2a7d528acb · 2012-08-24T11:23:50.000-04:00
Fix by Colin Su.  Test by me, based on a test written by Petri Lehtinen.
diff --git a/Lib/email/generator.py b/Lib/email/generator.py
@@ -400,6 +400,8 @@ def _handle_text(self, msg):
         if msg._payload is None:
             return
         if _has_surrogates(msg._payload) and not self.policy.cte_type=='7bit':
+            if self._mangle_from_:
+                msg._payload = fcre.sub(">From ", msg._payload)
             self.write(msg._payload)
         else:
             super(BytesGenerator,self)._handle_text(msg)
diff --git a/Lib/test/test_email/test_email.py b/Lib/test/test_email/test_email.py
@@ -21,7 +21,7 @@
 from email.charset import Charset
 from email.header import Header, decode_header, make_header
 from email.parser import Parser, HeaderParser
-from email.generator import Generator, DecodedGenerator
+from email.generator import Generator, DecodedGenerator, BytesGenerator
 from email.message import Message
 from email.mime.application import MIMEApplication
 from email.mime.audio import MIMEAudio
@@ -1306,6 +1306,20 @@ def test_mangle_from_in_preamble_and_epilog(self):
         self.assertEqual(len([1 for x in s.getvalue().split('\n')
                                   if x.startswith('>From ')]), 2)
 
+    def test_mangled_from_with_bad_bytes(self):
+        source = textwrap.dedent("""\
+            Content-Type: text/plain; charset="utf-8"
+            MIME-Version: 1.0
+            Content-Transfer-Encoding: 8bit
+            From: aaa@bbb.org
+
+        """).encode('utf-8')
+        msg = email.message_from_bytes(source + b'From R\xc3\xb6lli\n')
+        b = BytesIO()
+        g = BytesGenerator(b, mangle_from_=True)
+        g.flatten(msg)
+        self.assertEqual(b.getvalue(), source + b'>From R\xc3\xb6lli\n')
+
 
 # Test the basic MIMEAudio class
 class TestMIMEAudio(unittest.TestCase):
diff --git a/Misc/ACKS b/Misc/ACKS
@@ -1020,6 +1020,7 @@ Ken Stox
 Dan Stromberg
 Daniel Stutzbach
 Andreas Stührk
+Colin Su
 Pal Subbiah
 Nathan Sullivan
 Mark Summerfield
diff --git a/Misc/NEWS b/Misc/NEWS
@@ -29,6 +29,9 @@ Core and Builtins
 Library
 -------
 
+- Issue #15249: BytesGenerator now correctly mangles From lines (when
+  requested) even if the body contains undecodable bytes.
+
 - Issue #15777: Fix a refleak in _posixsubprocess.
 
 - Issue ##665194: Update email.utils.localtime to use datetime.astimezone and
