Hey- [EMAIL PROTECTED] wrote: > Hi list, > > the constructor of the new GML2 parser gets a featureType parameter. > > Does this mean that if a GML file contains multiple featureTypes, that I > have to call read multiple times, and append the results myself? Will this > be as efficient as doing the parsing in a single round?
Yeah, this is because up until this point, vector formats have been used to read/write collections of features, and I think of collections of features as layers. I'd like to continue thinking that way (because I think it works). But I know there are people who are interested in making WFS requests for multiple feature types (which I like to think of as layers). I really want to get WFS working for the most common cases. And I want it to be solid (with good support for transactions and filtered queries). I've made a boatload of changes in the topp/wfs sandbox - including changes to the GML & Filter code. My spare energy will be focussed on getting this code (the WFS protocol and related GML & Filter changes) in to the trunk. Modifying the existing GML code is a bit at odds with that. I also know that it will make sense to have the protocol be used in the absence of a layer. And of course we want to support our OGC protocol power user(s). Bart, can you describe (perhaps again) where you want to make a single request for multiple feature types. I'm sure it is done all the time, I just usually see a way that I think it simpler, so I'm curious to hear more use cases. Tim > > Best regards, > Bart > > _______________________________________________ > Users mailing list > [email protected] > http://openlayers.org/mailman/listinfo/users -- Tim Schaub OpenGeo - http://opengeo.org Expert service straight from the developers. _______________________________________________ Users mailing list [email protected] http://openlayers.org/mailman/listinfo/users
