The evidence I have for this is a bit "vague" being more of an impression that processing is faster, but it seems that evo runs one filter on a message, then the next, and so on. If you can almalgamate rules that do similar functions (i.e., copy messages to the same directory) into one compound rule, they seem to run faster as they are dealing with fewer, albeit more complex filters.
It would be nice to know exactly how the filtering mechanism works for this. To re-iterate, the majority of delays (after implementing stop rules per filter etc) are outside evo's control, particularly online spam checking and these are areas its not really possible to do much about if you need the functionality (and I certainly do!). BillK On Wed, 2005-08-24 at 12:21 -0400, fire-eyes wrote: > On Wed, 2005-08-24 at 09:06 +0800, W.Kenworthy wrote: > > 3. use a compound filter rather than separate ones (this assumes there > > is less overhead doing this - subjectively it does seem quicker) > > Thanks for the tips. The above one is the only one I don't really > understand, can you elaborate a bit? > -- gentoo-user@gentoo.org mailing list
