Closed jtroberts closed 3 years ago
Couple potential issues:
How about this... what if the oe_configure_remote_layers.py
script also created the mapfiles for remote layers (if create_mapfile==true
). That seems consistent with what the oe_configure_reproject_layer.py
script does. Then the oe_configure_layer.py
script would continue to create mapfiles as it was already doing for all XML layer configuration files. We would not deploy a layer configuration file for a layer that is found remotely... and if we do, then that's our bad and the oe_configure_layer.py
script would simply overwrite with a new mapfile (reading from MRF). We could log a SigEvent warning for that overwrite situation perhaps.
@mcechini I made the modifications as you suggested. One thing to note is that this will negate the fix for ONEARTH-611 for any layers that aren't included in the list of remote layers (since the solution to that issue is to use a WMTS layer source).
oe_configure_layer will use the layer's WMTS source (if it exists) for WMS mapfiles by leveraging oe_configure_reproject_layer. Otherwise it will still point to MRF files on disk. This plays a lot nicer with time snapping.
Possibly more changes to come in ONEARTH-665 if we need a more thorough solution.