The examples need to be improved, and in particular need to be expanded to include examples specific to both raster and tabular datasets. Right now it seems the examples convey the idea that the backend will do transformations from query types/values to the index types/values, but this is not the case: it is up to the client to ensure their query parameters match the types of the index values, which may be different than the data value types (particularly in the case of raster data).
It may be that effective examples will only come after a real PoC is implemented, to demonstrate what requests/responses must practically look like.
The examples need to be improved, and in particular need to be expanded to include examples specific to both raster and tabular datasets. Right now it seems the examples convey the idea that the backend will do transformations from query types/values to the index types/values, but this is not the case: it is up to the client to ensure their query parameters match the types of the index values, which may be different than the data value types (particularly in the case of raster data).
It may be that effective examples will only come after a real PoC is implemented, to demonstrate what requests/responses must practically look like.