|
From: Saul F. <Sau...@st...> - 2005-10-26 14:57:22
|
Also, one way to make this faster, Alex, would be create a view based on your query. Then simply call for the view in the getmap, rather than the source table. Of course, this will only work if there are a small number of values for the discriminating attribute in your query, since you don't want to create 500 views or something. --saul On Tue, 2005-10-25 at 23:46 -0400, db...@op... wrote: > Alex, > > 1. The error your report looks like a bug. Is there any more > information in the log? How much memory are you giving to Geoserver? > > 2. The way WMS and WFS handle <filters> are different. > > For WFS, they do what you want them to do -- they always pass the > filter off to the datastore. That means it gets converted into the SQL > WHERE clause. > > For WMS, it handles <filters> differently because there could be > multiple filters and else clause (they renderer features that were not > caught by any of the other rules). > > So, it makes a request for all the features that intersect the view > rectangle. It also makes a set of Rules that are valid for the current > scale range. The main loop looks like this: > > For each Feature in the view rectangle (Datastore processes this) > For each Rule (all handled by java) > IF the Feature satisfies the <Filter> render it > If no Rules were 'hit', then render with the <Else> rule > > One of the ways of speeding this up will be (if there are no ELSE > rules or empty-filter rules) to <Or> together all the filters and send > that off to the Datastore. For datastores like PostGIS, this will be > quick if there are only a FEW features in the view box are rendered > (ie. only show highways in a giant roads dataset). If most features > are rendered (ie. show all the roads in a road dataset) it will be a > bit slower because it is processing each filter twice -- once by the > datastore and once by the <Rule> processor. > > I've been thinking about doing this for a long time, but most people > tend to render most of the features, so its not actually going to make > this faster for most people. In fact, its likely to make things a > little slower. > > I'll ask the geotools people about it, > > dave > > > ---------------------------------------------------------- > This mail sent through IMP: https://webmail.limegroup.com/ > > > ------------------------------------------------------- > This SF.Net email is sponsored by the JBoss Inc. > Get Certified Today * Register for a JBoss Training Course > Free Certification Exam for All Training Attendees Through End of 2005 > Visit http://www.jboss.com/services/certification for more information > _______________________________________________ > Geoserver-users mailing list > Geo...@li... > https://lists.sourceforge.net/lists/listinfo/geoserver-users |