https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6576
John Hardin <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |[email protected] --- Comment #12 from John Hardin <[email protected]> 2011-05-04 04:26:13 UTC --- (In reply to comment #10) > (In reply to comment #5) > > > - The rule does what it's supposed to do: tag a borked header directly > > connecting to an MTA. > > There is nothing wrong with missing angle brackets in an e-mail address > which does not also carry a display name. See RFC 5322 section 3.4.: As has been stated before, SA is not an RFC Compliance Audit engine. Otherwise the postmaster.rfc-ignorant.org BL would have a high score. A bare unbracketed email address in the To: header appears often enough in spam vs. ham that it's useful to look for. > I strongly agree that a score: > score TO_NO_BRKTS_DIRECT 4.299 3.483 4.299 3.483 > is way too high for a perfectly legitimate mail. > > Suggesting to tame down at least that score. Reopening. I've been working for a while to reduce its FPs. The OP has a network topology and policy that makes unwanted FPs on direct-to-MX and dynamic IP mail inevitable. I made some suggestions in private email for working around it. Perhaps we need a way to set the maximum score that can be assigned to any one rule by the generator? That rule also seems to have dropped into the abyss in recent masscheck runs. I'm going to have to see what happened to it... (In reply to comment #11) > Btw, there were complaints about TO_NO_BRKTS_DIRECT... on a mailing list > back in January 2011. I don't find any complaints back then, just a query about what the description meant. -- Configure bugmail: https://issues.apache.org/SpamAssassin/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug.
