pairs of DimVector do not reflect lookup values

[asinghvi17]

Not sure if this should happen or not, but see the following MWE:

julia> DimVector(rand(10), (X(16:25),)) |> pairs
pairs(::DimVector{Float64, Tuple{X{DimensionalData.Dimensions.Lookups.Sampled{Int64, UnitRange{Int64}, DimensionalData.Dimensions.Lookups.ForwardOrdered, DimensionalData.Dimensions.Lookups.Regular{Int64}, DimensionalData.Dimensions.Lookups.Points, DimensionalData.Dimensions.Lookups.NoMetadata}}}, Tuple{}, Vector{Float64}, DimensionalData.NoName, DimensionalData.Dimensions.Lookups.NoMetadata})(...):
  1  => 0.151612
  2  => 0.400987
  3  => 0.928486
  4  => 0.245337
  5  => 0.747621
  6  => 0.327087
  7  => 0.781531
  8  => 0.0161408
  9  => 0.930245
  10 => 0.656637

I would naively expect that the LHS values on the pairs would be from 16 to 25 (that's what OffsetArrays does).

[asinghvi17]

I guess that indexing the dimvector works with one based indexing, so there is a good reason to have the pairs work the way they are. Feel free to close this issue then :D (though at least it's searchable now)

[felixcremer]

If you want the dimensions in there you can have a look at DimIndices.

This would give you the keys with the Dimension information.

[rafaqz]

I don't think much thought went into it honestly

[rafaqz]

We can wrap it with At for 16:25 to work as indices, so it could be:

  At(16)  => 0.151612
  At(17)  => 0.400987
  3  => 0.928486
...
  At(25) => 0.656637

But it seems kinda weird, and close to what DimSelectors gives you but less useful.

[asinghvi17]

Yeah, as long as direct indexing with (idx, val) in pairs(dimarray) works (which it looks like it does) then it's probably fine.