[Geoprisma-users] FeatureServer, OL, GP ack!

[Stephen Woodbridge]

So you said tinyOWS was not supported in GP yet.
So I installed FS and started reading your past threads on implementing 
FS with OL, then looked at http://dev4.mapgears.com/bdga-fs/

Now my head is ready to explode :) And this was easier then adding 
support of tinyOWS?
ok I know that is rhetorical, but I'm just saying ...

Anyway, this may be a little OT, but my GP use case is a web based GIS 
systems or something close to that. And I have all data layers loaded in 
postGIS, I can create a Mapserver mapfile to display these, I already 
have one displaying them from the shapefiles. And I'm trying to figure 
out how to integrate FS and how this fits into the bigger design.

In OL, I will likely have Mapserver (WMS? or not) base layer, and Google 
base layer. Then I need to be able to select one of the layers to be 
editable (I'm assuming only one can be editable at a time) and it will 
then get overdrawn via FS so edits can take place.

I'm not sure the best way to select the layer for editing. It probably 
means hacking OL.layerswitcher so that the layers are in a radio button 
field that also selects the layer as editable and only one can be on at 
a time. This probably means the select control needs to be notified 
maybe it is listening for a layer switch change event.

Seems like the draw/edit tools needs to be selected based on the layer 
type polygon, line, point.

So I guess I need to pick apart the bdga-fs demo. And ideas on how to 
find out what the layer type is?

And finally (or not) GP needs to be wrapped around all of this.

I think someone mentioned that a future enhancement was to be able to 
upload a shapefile and install it under GP. I assume that would get 
loaded into postGIS, then a mapfile generated, FS cfg fragment would be 
generated and added to featureserver.cfg or something along these lines. 
Where does this stand in your plans? I'm assuming this is all initiated 
from an OL application so when the server is ready with the new data, it 
response to OL and the code there will dynamically add a new OL layer to 
the map and refresh the page. Also any GP config files also need to be 
updated. Since a lot of stuff needs to get changed, I might make sense 
that the yours needs to reload the browser page before the changes are 
visible instead of trying to dynamically update the page.

Thoughts, examples, help appreciated.

-Steve