parser: Ignore CFWS in In-Reply-To, References headers

RFC2822 states that [1] a comment or folding white space is permitted to
be inserted before or after a msg-id in in the Message-ID, In-Reply-To
or References fields. Allow for this.

[1] https://tools.ietf.org/html/rfc2822#section-3.6.4

Signed-off-by: Stephen Finucane <[email protected]>
Reviewed-by: DJ Delorie <[email protected]>
Closes: #399

Author: stephenfin

patchwork/parser.py

@@ -31,6 +31,7 @@
 from patchwork.models import State
 
 
+_msgid_re = re.compile(r'<[^>]+>')
 _hunk_re = re.compile(r'^\@\@ -\d+(?:,(\d+))? \+\d+(?:,(\d+))? \@\@')
 _filename_re = re.compile(r'^(---|\+\+\+) (\S+)')
 list_id_headers = ['List-ID', 'X-Mailing-List', 'X-list']
@@ -502,19 +503,15 @@ def find_references(mail):
 
     if 'In-Reply-To' in mail:
         for in_reply_to in mail.get_all('In-Reply-To'):
-            r = clean_header(in_reply_to)
-            if r:
-                refs.append(r)
+            ref = _msgid_re.search(clean_header(in_reply_to))
+            if ref:
+                refs.append(ref.group(0))
 
     if 'References' in mail:
         for references_header in mail.get_all('References'):
-            h = clean_header(references_header)
-            if not h:
-                continue
-            references = h.split()
+            references = _msgid_re.findall(clean_header(references_header))
             references.reverse()
             for ref in references:
-                ref = ref.strip()
                 if ref not in refs:
                     refs.append(ref)
 

patchwork/tests/test_parser.py

@@ -74,6 +74,7 @@ def _create_email(
     sender=None,
     listid=None,
     in_reply_to=None,
+    references=None,
     headers=None,
 ):
     msg['Message-Id'] = msgid or make_msgid()
@@ -84,6 +85,9 @@ def _create_email(
     if in_reply_to:
         msg['In-Reply-To'] = in_reply_to
 
+    if references:
+        msg['References'] = references
+
     for header in headers or {}:
         msg[header] = headers[header]
 
@@ -97,12 +101,20 @@ def create_email(
     sender=None,
     listid=None,
     in_reply_to=None,
+    references=None,
     headers=None,
 ):
     msg = MIMEText(content, _charset='us-ascii')
 
     return _create_email(
-        msg, msgid, subject, sender, listid, in_reply_to, headers
+        msg,
+        msgid,
+        subject,
+        sender,
+        listid,
+        in_reply_to,
+        references,
+        headers,
     )
 
 
@@ -1290,30 +1302,34 @@ def log_query_errors(execute, sql, params, many, context):
 
     def test_duplicate_patch(self):
         diff = read_patch('0001-add-line.patch')
-        m = create_email(diff, listid=self.listid, msgid='[email protected]')
+        m = create_email(diff, listid=self.listid, msgid='<[email protected]>')
 
         self._test_duplicate_mail(m)
 
         self.assertEqual(Patch.objects.count(), 1)
 
     def test_duplicate_comment(self):
         diff = read_patch('0001-add-line.patch')
-        m1 = create_email(diff, listid=self.listid, msgid='[email protected]')
+        m1 = create_email(
+            diff,
+            listid=self.listid,
+            msgid='<[email protected]>',
+        )
         _parse_mail(m1)
 
         m2 = create_email(
             'test',
             listid=self.listid,
-            msgid='[email protected]',
-            in_reply_to='[email protected]',
+            msgid='<[email protected]>',
+            in_reply_to='<[email protected]>',
         )
         self._test_duplicate_mail(m2)
 
         self.assertEqual(Patch.objects.count(), 1)
         self.assertEqual(PatchComment.objects.count(), 1)
 
     def test_duplicate_coverletter(self):
-        m = create_email('test', listid=self.listid, msgid='[email protected]')
+        m = create_email('test', listid=self.listid, msgid='<[email protected]>')
         del m['Subject']
         m['Subject'] = '[PATCH 0/1] test cover letter'
 
@@ -1322,6 +1338,52 @@ def test_duplicate_coverletter(self):
         self.assertEqual(Cover.objects.count(), 1)
 
 
+class TestFindReferences(TestCase):
+    def test_find_references__header_with_comments(self):
+        """Test that we strip comments from References, In-Reply-To fields."""
+        in_reply_to = (
+            '<[email protected]> (message from\n'
+            ' liqingqing on Thu, 1 Apr 2021 16:51:45 +0800)'
+        )
+        email = create_email('test', in_reply_to=in_reply_to)
+
+        expected = ['<[email protected]>']
+        actual = parser.find_references(email)
+
+        self.assertEqual(expected, actual)
+
+    def test_find_references__duplicate_references(self):
+        """Test that we ignore duplicate message IDs in 'References'."""
+        message_id = '<[email protected]>'
+        in_reply_to = (
+            '<[email protected]>'
+        )
+        references = (
+            '<[email protected]>\n'  # noqa: E501
+            ' <CAE68AUOr7B5a2QvduJhH0kEHPi+sR9X3qfrtumgLxT1BK4VS+Q@mail.gmail.com>\n'  # noqa: E501
+            ' <[email protected]>\n'
+            ' <[email protected]>\n'
+            ' <[email protected]>'
+        )
+        email = create_email(
+            'test',
+            msgid=message_id,
+            in_reply_to=in_reply_to,
+            references=references,
+        )
+
+        expected = [
+            '<[email protected]>',
+            '<[email protected]>',
+            '<[email protected]>',
+            '<CAE68AUOr7B5a2QvduJhH0kEHPi+sR9X3qfrtumgLxT1BK4VS+Q@mail.gmail.com>',  # noqa: E501
+            '<[email protected]>',
+        ]
+        actual = parser.find_references(email)
+
+        self.assertEqual(expected, actual)
+
+
 class TestCommentCorrelation(TestCase):
     def test_find_patch_for_comment__no_reply(self):
         """Test behavior for mails that don't match anything we have."""

releasenotes/notes/issue-399-09d6f17aa54b14b2.yaml

@@ -0,0 +1,11 @@
+---
+fixes:
+  - |
+    Comments and whitespace are now correctly stripped from the ``Message-ID``,
+    ``In-Reply-To``, and ``References`` headers. One side effect of this change
+    is that the parser is now stricter with regards to the format of the
+    ``msg-id`` component of these headers: all identifiers must now be
+    surrounded by angle brackets, e.g. ``<[email protected]>``. This is
+    mandated in the spec and a review of mailing lists archives suggest it is
+    broadly adhered to. Without these markers, there is no way to delimit
+    ``msg-id`` from any surrounding comments and whitespace.