http://issues.apache.org/SpamAssassin/show_bug.cgi?id=4776





------- Additional Comments From [EMAIL PROTECTED]  2006-02-02 21:46 -------
I'm +1 on the change, or at least the idea.  If we don't pluginize check, it
will be impossible for people to test new models of short-circuiting/priorities
without modifying Mail::SpamAssassin::PerMsgStatus.

For example, consider a decision-tree approach, where the check implementor
plugin picks individual rules of different types to run, one-by-one, based on
the results of previous rules.   In that situation, the largest negative score
of a rule type etc. is irrelevant; the only way to do it is to replace the
entire check() function's operation somehow.

by the way, I'd be happy to include the current check() code as a default
implementation that's used if no plugin implements check_main(), hence avoiding
the situation where no active plugin implements that method; I can agree
that would be a bad situation and one worth avoiding.


But this pluginized-check change is one that's been planned for a very long 
time.



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

Reply via email to