On Mon, 26 Jan 2004, Regis Wilson wrote: > As was noted before, a negative test will false-positive when there is > no X-O-IP header.
Right, but I did resubmit the rule with an [if-unset] clause added. If you missed that message, I can resend you a copy off-list, but the gist is to simply append "[if-unset 0.0.0.0]" which will satisfy the test if the header is not present. > Thus, (compound rule omitted) I prefer to avoid the compound rule in this case, since a) it _can_ be avoided, b) the single rule is less work for the mail server (my mail servers are quite powerful systems, but they process a lot of mail and adding unnecessary tests is starting to become an issue for me), and c) it's easier for a human to read and understand the single rule, which makes for easier maintenance. In other cases, a compound rule is the only reasonable approach, of course ... > (My code is more strict about the format of the IP address, but it's > the same thing...) Right. I had contemplated an almost identical stricter IP address check, but decided that for the purpose at hand it really isn't necessary. This isn't being used to determine where the message will be sent, after all, just whether or not the message has this header's content properly formatted if it's present. The simpler check for a dotted-quad with optional surrounding brackets is (in my opinion, of course), easier to read and understand, and likely less work for the software to check against (though I have no benchmarking tests to support that hypothesis). The readability is rather important to me, in cases where others will need to understand what I've written. -- ---------------------------------------------------------------------- Sylvain Robitaille [EMAIL PROTECTED] Systems analyst / Postmaster Concordia University Instructional & Information Technology Montreal, Quebec, Canada ---------------------------------------------------------------------- ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
