My ansewer has been to have the tests need to be unique (not subsets of another test) for this to work. I've been burned by the testfailed with the same substring in two different tests... Here's a case where a search for a preceding space would be nice (although then the test would be first...)
Perhaps you should use: Spamhaus-SBL-IP4R Spamhaus-SBL-COMBO I would imagine all of the other options do work with testsfailed (ENDSWITH, etc,) They just they aren't very useful . ---------- Original Message ---------------------------------- From: "Kami Razvan" <[EMAIL PROTECTED]> Reply-To: [EMAIL PROTECTED] Date: Fri, 26 Nov 2004 12:04:38 -0500 >Hi; > >I don't think with the current TESTSFAILED option one can uniquely identify >a single test or can we? > >Example: > >I have broken down all tests into combination filters with a naming >convention. > >IP4R-something > >COMBO-IP4r-something > >Now I can write combo filters that are: > >TESTSFAILED END CONTAINS COMBO- > >Or > >TESTSFAILED END CONTAINS COMBO-IP4R > >But I can not say > >TESTSFAILED END <is the exact name of one of the failed test> >IP4R-something > >without triggering the COMBO-IP4R-something > >I guess the question is one can not detect a failed test by its exact name. >In the manual the only option for TESTSFAILED is either CONTAINS or >NOTCONTAINS. Can one find out if a TEST's name is among the tests that have >failed but as the exact name of the test? > >Regards, >Kami > > --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type "unsubscribe Declude.JunkMail". The archives can be found at http://www.mail-archive.com.
