You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Not sure if there is still room to change a lot here, but while looking into vector cube issues I observed this:
filter_bbox supports both raster and vector cubes
filter_spatial supports only raster cubes and filter_vector support only vector cubes, but both work almost identical (assuming relation="intersect" in filter_vector). Is there some opportunity to unify this better, e.g. allow vector cubes in filter_spatial?
I think at least we should have better cross-referencing (documentation-wise) between these processes (there is no documentation link between filter_spatial and filter_vector)
The text was updated successfully, but these errors were encountered:
I considered it, but based on the relation parameter I found it akward.
You'd need to assume the Pixels are points at the pixel center, but I'm not sure whether you'd want to implement all the allowed relation types that you have for vectors also for rasters. What happens if you only support intersection for raster and all other relation types for vector? Confusing. filter_bbox was merged as it doesn't support the relation types and is much simpler in that regard.
But indeed, crosslinking between them would be good.
Not sure if there is still room to change a lot here, but while looking into vector cube issues I observed this:
filter_bbox
supports both raster and vector cubesfilter_spatial
supports only raster cubes andfilter_vector
support only vector cubes, but both work almost identical (assuming relation="intersect" in filter_vector). Is there some opportunity to unify this better, e.g. allow vector cubes in filter_spatial?I think at least we should have better cross-referencing (documentation-wise) between these processes (there is no documentation link between filter_spatial and filter_vector)
The text was updated successfully, but these errors were encountered: