| Upendra, Well, anything is possible, but you are right to note that there are a number of ways that GML can be more complex than simple geometry models, and exposing that complexity to the user for manipulation gets complex along with the model. I think that, with a proper generic topology model in geotools and some tools in udig doing GML topology via WFS would be possible too. But note that there is some core work to be done before the WFS part even becomes possible. I think that for multiple-geometry schemas it would not be too hard to allow people to choose which geometry element they want rendered and enable for editing. I think that in general deeply nested feature types could be tricky, but not impossible to expose to the user. Most of the infrastructure for holding and working with those schemas already exists in geotools, but there is some design and thought required on how to expose it to users (particularly how to do so without confusing the default case). In answer to your final question, yes, it is possible to do a general purpose client without knowing the application schema beforehand, as udig already does that. It reads the application schema and configures its parser appropriately. However more complicated cases, particularly some of the things in GML3, will require extension to support. And the $1M question is: if the use of the extended features in GML3 is expected to be infrequent, is it worth spending the effort to support them? P On 6-Oct-06, at 2:29 PM, Upendra wrote: Hello everybody, |
_______________________________________________ User-friendly Desktop Internet GIS (uDig) http://udig.refractions.net http://lists.refractions.net/mailman/listinfo/udig-devel
