Re: [SAtalk] Meta tests

On Thu, Dec 26, 2002 at 11:56:29PM -0500, Theo Van Dinter wrote: > So in 2.50 now, naming doesn't matter now for meta meta dependencies; > the code will figure out what order to run the tests in, including > circular dependencies, and do the right thing. :) Way cool, thanks. -Jeremy --

Re: [SAtalk] problem with user_pref level (possible bug?)

On Thu, Dec 26, 2002 at 11:04:12AM -0600, Mark Merner wrote: > for a temporary fix I could comment out the spam15 level, but it should > really be changed to something that checks for Status:Yes and 15 *. Let > me know if this doesn't make any sense. Well, it makes sense to me, but there's no bug

Re: [SAtalk] Meta tests

On Thu, Dec 26, 2002 at 06:27:47PM -0500, Theo Van Dinter wrote: > Ok, I cleaned up the do_meta code a little bit and added in the > "strategic" sort. :) I first tried changing the line my @tests = keys %{$self->{conf}{meta_tests}}; to just sort the keys, but that explodes badly -- you pret

Re: [SAtalk] Meta tests

On Thu, Dec 26, 2002 at 06:27:47PM -0500, Theo Van Dinter wrote: > Ok, I cleaned up the do_meta code a little bit and added in the > "strategic" sort. :) Ok again, after fussing around with do_meta tonight, I figured out that it wouldn't be hard to put in the full dependency algorithm, so I coded

Re: [SAtalk] Meta tests

On Thu, Dec 26, 2002 at 05:45:31PM -0500, Theo Van Dinter wrote: > I'll look at the code and see what I can come up with. I'd think, as you > say, a strategically placed sort() would do the trick for the time being. Ok, I cleaned up the do_meta code a little bit and added in the "strategic" sort.

[SAtalk] SA 2.50-cvs

I'm very happy with the bayes and other additions for 2.50-cvs.. I've been running it for the past 2 weeks... and I have had only 3 false positives (which where spam, but from sources I wanted email from). Can't wait for the release! --- This

Re: [SAtalk] Meta tests

On Thu, Dec 26, 2002 at 05:09:57PM -0500, Jeremy Nixon wrote: > I suspect my problem is related to the order in which meta rules are > evaluated, or the "legality" of nesting them. Is there some limit to > the depth to which meta rules can be nested inside other meta rules? > I had one like "meta

[SAtalk] Meta tests

So I'm having some difficulty grasping some stuff about "meta" rules. Maybe now that I've got >350 lines of local spamassassin config, it's time for me to delve into the source and join the devel list or something, but let's see if I can figure this out. I suspect my problem is related to the ord

RE: [SAtalk] Razor2 error

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wednesday, December 25, 2002 12:40 PM Theo Van Dinter wrote: > On Wed, Dec 25, 2002 at 09:34:38AM -0600, Mike Loiterman wrote: >> Is the razor2 server down? I'm getting a lot of these messages in >> my maillog: razor

[SAtalk] problem with user_pref level (possible bug?)

I've been using SA for about 2 weeks now. Great Program, but I think I may have found a problem. (I'm running sendmail+procmail+SA2.43, NOT spamc/spamd) I upped one user's spam level to 25, because they receive ads that go in to a newspaper. Here is the problem. If a email gets marked with a leve

Re: [SAtalk] X-Rot version

On Wed, 25 Dec 2002 the voices made Vivek Khera write: VK> > "TLS" == Tony L Svanstrom <[EMAIL PROTECTED]> writes: VK> VK> TLS> I want to add one thing: SA could actually be a weakness... VK> VK> TLS> As one rule among many this sure spamsign might not be enough to VK> TLS> tag the e-mail as