Naser Mahfouz
Naser Mahfouz
@bartgol this PR is ready, three items for your input, please :) it's been only a few weeks, but I am 100% rusty on details, so please bear with me...
@bartgol, if we end up doing this, we should implement it in a composable fashion so that it can easily be applied to all sorts of stats (min, max, weighted...
> this feature should not be hard to add. I wonder if we should make it into a new "remapper", so that we can piggy back on the existing infrastructure...
Yeah, I agree there are perf implications of the generic approach. I didn't notice that the request was specific for _ALL_ variables in a given file. While the remapper idea...
@bartgol I think the fail is to do with comparison against baselines? If so, that's expected. This PR takes away eight fields.
> What does "planning accordingly" mean? How do they get those fields now? The deleted eight diagnostics (potentially with slightly different answers) can be requested with all-or-nothing diagnostic term `AeroComCldTop`....
@bartgol did you confirm that the comparison tests are failing due to missing fields? Or is it something else?
Oh no, no task! I just wanted a second pair of eyes to check. We also shouldn't merge until Ben gets a lot of time to review this.
This PR is ready for merging, but I will give Ben another week or so since it's low priority :)
WIP status: will need rebasing.