From d0e7c31dc69f320c724f7d3c081639962e3b7ab6 Mon Sep 17 00:00:00 2001 From: Linus Nordberg Date: Thu, 11 Aug 2022 10:19:23 +0200 Subject: new file; update mailman --- wiki/mailman.md | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'wiki/mailman.md') diff --git a/wiki/mailman.md b/wiki/mailman.md index c9b5c2b..fa73770 100644 --- a/wiki/mailman.md +++ b/wiki/mailman.md @@ -46,3 +46,23 @@ Type "help", "copyright", "credits" or "license" for more information. >>> email.utils.getaddresses([s]) [('12 45678901234567890123456789012345678901 3456789012345678901234 678901234', 'testing@lists.sigsum.org')] ``` + + +2022-08-05 in #mailman@Libera.Chat +``` + hi all, have anyone of you experienced erroneous moderation due to + implicit destination when posting with To: "long string with spaces, 76 + chars will do" ? [09:05] + i tracked it down to mailman/src/mailman/rules/implicit_dest.py and + verified that mail.utils.getaddresses() at least doesn't get things + wrong, so i guess it might be the msg.get_all(header, []) that's borken + [09:07] + now, to debug this what's a decent option? simply editing + /usr/lib/python3/dist-packages/mailman/rules/implicit_dest.py (i'm on + debian11, with the mailman3 package installed) and adding print()'s? or + what do you suggest? [09:10] + to further incite you to care about this, the case is a real one: the + dmarc mitigation strategy of rewriting the address part of From: to the + list address and composing a fullname part indiciating whoe the sender + is can generate a From: header which is then used as To: in a reply +``` -- cgit v1.2.3