http://bugzilla.spamassassin.org/show_bug.cgi?id=4400
------- Additional Comments From [EMAIL PROTECTED] 2005-06-13 10:52 ------- Well, I'm not going to raise a ruckus if it doesn't go into 3.1.0, though I think it should (and I'll tweak my 3.1.0 install so it does there) since I see it as obviously equivalent code with a significant upside (1/3 less writing on inserts and updates). I suppose I may have been a little unclear before about it's testing status---I've been running like this for a week, and before I was having exactly the problem that the additional index present in HEAD are intended to solve (sequential scans for every lookup). FWIW, I'm going to attach a short script that others could run that will test both setups, albeit against tokens that are in my db. I'm also going to attach the output of running that script (on my bayes db, with about 46K messages and 1.3M tokens), and it's easy to see that the exact same plans are being generated in both places. Seriously, this is a big win for any postgresql installations that see any measure of activity because it's lightening the write load on the disk subsystem, and, perhaps only as a result of my long use of PostgreSQL, I believe it is absolutely zero-risk. ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.
