> - the same column can contain geometries with different spatial > reference system too. I know it sounds crazy, but as far as I remember > OGC WFS makes sure that you can handle that case as well... mixed > with the fact that in GeoServer you have to declare a SRS anyways > (we have no way to support a srs-less dataset at the moment) > it means the code cannot just check if the native SRS is null > (which is what I would expect from a mixed SRS geometry column). > Btw, Justin, can you confirm the above or it's just my memory failing > me?
The spec definitely does not prohibit it. But the CITE tests only ensure we can handle multiple geoms in the same referencing system. > > So it seems I can pass down a generalization distance only if > there is no reprojection going on and the native srs has not > been tampered with? > Alternatively I guess I could pass down a generalization distance > in the rendering SRS, and pass down the rendering CRS as well. > It will be the job of the datastore to figure out if the generalization > distance need to be adjusted given the native SRS of the data, or not. > Opinions? > > Cheers > Andrea > -- Justin Deoliveira OpenGeo - http://opengeo.org Enterprise support for open source geospatial. ------------------------------------------------------------------------------ _______________________________________________ Geoserver-devel mailing list Geoserver-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geoserver-devel
