Dr Tom August

Seeing as I can't remember the original issue your solution sounds fine. But just watch out for times where the na attibute might cause problems.

It think you could make this work. My issue would be that what is returned is not a zoonworkflow object as it does not meet the criteria expected by other...

Smashed it! ;) 

_drops microphone and walks off stage_

`OccurrenceTable` seems best to me. I also think that this should work too: ``` R df

I like this idea. This is what I needed for the appify module. If we think output modules will either work by aggregating or not (i.e. not both) then we...

A catch all might be ``` R PredictNewArea(newArea = ras) ``` where ras can be a raster (user provided), or a call to a covariate module: ``` R PredictNewArea(newArea =...

This has links with #121 in terms of the code required for the solution

I agree 1 is the ideal solution. This would also require a review of existing models to ensure they work with `NA` values and addition of `NA` values to the...

Yes, I mean throw a sensible error or remove NAs with warning