http://bugzilla.spamassassin.org/show_bug.cgi?id=3616





------- Additional Comments From [EMAIL PROTECTED]  2005-01-11 11:16 -------
I like Scott's suggestion.  But it applies to SPF in general, not just neutral
SPF records.
E.g. a +all should be scored differently than a +ip4, just as a 
?all should be scored differently than a ?ip4, and 
+all should be scored differently than a +ip4...
  Because it's more ambitious, perhaps a separate bug would be better, as this
bug has a fix already *that looks checkinable to me* ?  

Also regarding Scott's suggestion: Perhaps the following knobs (defined to work
on HELO and envelope-from) would be sufficient (with my guesstimate as to good
start/likely optimized values - Yes, I'm guessing that a +all match is stronger
spamsign than a -all match. <sic>)
 score SPF_ALL_PASS 2.5 
 score SPF_ALL_NEUTRAL 1.5 
 score SPF_ALL_FAIL 2 
Perhaps, instead of 3, a single add'l knob would be sufficient:
 score SPF_ALL 2 

Also here's a suggested documentation change (in pseudo-diff syntax) that makes
the comment accurately reflect what the SPF specs actually require:
in lib/Mail/SpamAssassin/Plugin/SPF.pm:
-     # SPF HELO-checking variant.  This isn't really SPF at all ;)
+     # SPF HELO-checking variant.  SPF requires this be valid too.




------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

Reply via email to