Rafael Schouten

No worries, please add to Tiems issue or here with any problems you have with functionality or docs. A comparison to xarray is good to improve things here! Mostly @sethaxen...

We could support `mapslices` too if you want to add it. But I think eachslice is favoured these days?

I think we also need the version with a function as first argument. And also is count really in that list? Especially with the function? I thought it would need...

Oh right the dispatch on internals may have broken, sorry I should have realised that. We also just switched this to the main method in DimensionalData.jl https://github.com/meggart/DiskArrays.jl/blob/ccd3092ca6f32c2663ded1d74176aa9dd222a5ef/src/mapreduce.jl#L7-L12 Sorry @Alexander-Barth for...

Seems this has been pretty productive! good to have all those things working and tested. I think youre right - `sum`, `prod`, `any` and `all` are the real beneficiaries of...

DimStack is a totally custom creation (it's like a NamedTuple/Array Franken thing) so we can do what we want. It makes sense to add a replace! method. You may also...

But probably best just to do : ```julia foreach(layers(stack)) do A replace!(A, 0 => NaN) end ``` Wich generalises to : ```julia foreach(layers(stack), replacementvals) do A, val replace!(A, zero(val) =>...

Rethinking this it doesn't really make sense as DimStack can have mixed type layers. Doing it manually on layers is better.

Not much we can do with a broadcast. Maybe we need a dedicated `combine` function for this. ```julia combine(mean, groupby(A, Ti => year)) ``` We can store the time dim...

The problem is it's pretty ambiguous what `mean` means here. XArray hacks mean to do an inner mean and cat like you want. But that's not very Julian. If DimGroupbyArray...