[Python-checkins] [3.11] gh-106186: Don't report MultipartInvariantViolationDefect for valid multipart emails when parsing header only (GH-107016) (#107112)

Sun Jul 23 08:28:39 EDT 2023

https://github.com/python/cpython/commit/afa24d52b821c2020e0966f96a6205bca7db85e9
commit: afa24d52b821c2020e0966f96a6205bca7db85e9
branch: 3.11
author: Miss Islington (bot) <31488909+miss-islington at users.noreply.github.com>
committer: ambv <lukasz at langa.pl>
date: 2023-07-23T14:28:35+02:00
summary:

[3.11] gh-106186: Don't report MultipartInvariantViolationDefect for valid multipart emails when parsing header only (GH-107016) (#107112)

(cherry picked from commit c65592c4d6d7552fb6284442906a96a6874cb266)

Co-authored-by: htsedebenham <31847376+htsedebenham at users.noreply.github.com>

files:
A Lib/test/test_email/data/msg_47.txt
A Misc/NEWS.d/next/Library/2023-07-22-13-09-28.gh-issue-106186.EIsUNG.rst
M Lib/email/feedparser.py
M Lib/test/test_email/test_email.py

diff --git a/Lib/email/feedparser.py b/Lib/email/feedparser.py
index 97d3f5144d606..e400dc7fb89b0 100644
--- a/Lib/email/feedparser.py
+++ b/Lib/email/feedparser.py
@@ -189,7 +189,7 @@ def close(self):
         assert not self._msgstack
         # Look for final set of defects
         if root.get_content_maintype() == 'multipart' \
-               and not root.is_multipart():
+               and not root.is_multipart() and not self._headersonly:
             defect = errors.MultipartInvariantViolationDefect()
             self.policy.handle_defect(root, defect)
         return root
diff --git a/Lib/test/test_email/data/msg_47.txt b/Lib/test/test_email/data/msg_47.txt
new file mode 100644
index 0000000000000..bb48b47d96baf
--- /dev/null
+++ b/Lib/test/test_email/data/msg_47.txt
@@ -0,0 +1,14 @@
+Date: 01 Jan 2001 00:01+0000
+From: arthur at example.example
+MIME-Version: 1.0
+Content-Type: multipart/mixed; boundary=foo
+
+--foo
+Content-Type: text/plain
+bar
+
+--foo
+Content-Type: text/html
+<html><body><p>baz</p></body></html>
+
+--foo--
\ No newline at end of file
diff --git a/Lib/test/test_email/test_email.py b/Lib/test/test_email/test_email.py
index ca8212825ffca..2bb651609f572 100644
--- a/Lib/test/test_email/test_email.py
+++ b/Lib/test/test_email/test_email.py
@@ -3696,6 +3696,16 @@ def test_bytes_header_parser(self):
         self.assertIsInstance(msg.get_payload(), str)
         self.assertIsInstance(msg.get_payload(decode=True), bytes)
 
+    def test_header_parser_multipart_is_valid(self):
+        # Don't flag valid multipart emails as having defects
+        with openfile('msg_47.txt', encoding="utf-8") as fp:
+            msgdata = fp.read()
+
+        parser = email.parser.Parser(policy=email.policy.default)
+        parsed_msg = parser.parsestr(msgdata, headersonly=True)
+
+        self.assertEqual(parsed_msg.defects, [])
+
     def test_bytes_parser_does_not_close_file(self):
         with openfile('msg_02.txt', 'rb') as fp:
             email.parser.BytesParser().parse(fp)
diff --git a/Misc/NEWS.d/next/Library/2023-07-22-13-09-28.gh-issue-106186.EIsUNG.rst b/Misc/NEWS.d/next/Library/2023-07-22-13-09-28.gh-issue-106186.EIsUNG.rst
new file mode 100644
index 0000000000000..07fdcc96fa38a
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2023-07-22-13-09-28.gh-issue-106186.EIsUNG.rst
@@ -0,0 +1,3 @@
+Do not report ``MultipartInvariantViolationDefect`` defect
+when the :class:`email.parser.Parser` class is used
+to parse emails with ``headersonly=True``.

