https://bz.apache.org/SpamAssassin/show_bug.cgi?id=7735
--- Comment #67 from Loren Wilton <lwil...@earthlink.net> --- (In reply to Henrik Krohns from comment #65) > Not sure I follow, would it matter why a rule is unrun? People have been talking about meta rules that evaluate to something useful, and then "season" the result with a term from something like a net rule, that might not be run. If I understand them correctly, they are saying that they would like the result of that rule to be the result of it's expression, with the unrun rule zero. The result of this meta may not be zero, so the rule will end up scoring. If this meta were just set to zero because it was unrun, it likely would not produce the desired result. >From your comment #14: > Let's take this stock rule as example. > meta DIGEST_MULTIPLE RAZOR2_CHECK + DCC_CHECK + PYZOR_CHECK > 1 > This gets problematic even when considering disabled rules. If simply one of > these is not used/loaded, or lookup times out, it won't hit even if two of > them hit. So I claim that if razor and dcc run, but pyzor doesn't, the result should be: meta DIGEST_MULTIPLE RAZOR2_CHECK + DCC_CHECK + 0 > 1 Obviously the rule could still hit. -- You are receiving this mail because: You are the assignee for the bug.
