http://bugzilla.spamassassin.org/show_bug.cgi?id=3998





------- Additional Comments From [EMAIL PROTECTED]  2005-01-10 15:51 -------
> Hmmmm... okay, then, I revoke my -1.  It might be clearer and cleaner to have
> a single entry point (eval test) for all of the Habeas rules.

Agreed, but I coded it the way I did for four reasons. First, I was trying to
make as few changes to the base code as possible. Second, I wanted to be able to
write separate rules for the old Habeas style (Haiku + HUL) and the new (return
path + ACCREDIT); the easiest way to do that without breaking the existing rules
was to make separate entry points. Third, we'd like to open up the return path
encoding to other accreditation authorities, not just Habeas. Finally, I wanted
to make it obvious where to delete the old Haiku-based code when it comes to
that, again without breaking other existing rules.

I commented the code pretty carefully so newcomers would (I hoped!) understand
what was going on. 

> Can you please describe the automated testing a bit?  I just want to know
> that we're not giving out bonus points for something easily forged.

With SpamAssassin 3.0, all Habeas senders are verified by DNSWL lookup. There's
no opportunity for forging anything, either in the envelope or header. The
difference between the different Accreditation Levels is the nature of how
senders are added or remove from our whitelist, not how we authenticate them.



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

Reply via email to