Carlos Parada
Carlos Parada
> Gotcha, I'll see if I manage to put in some effort. > > This is in principle pretty simple right? (And simpler than implementing an adaptive scheme which tunes...
@kaandocal Are you still interested in getting this PR merged?
> We should probably get this PR out of the way. I did not pursue this much further because I had trouble getting it to work on my sampling problems...
I will note that I think this is the correct behavior, as `map` is a clearly functional construct and so should only be used with pure functions.
As for the current documentation, it's ambiguous, but I actually lean towards the reading of it as assuming purity. "Transform collection `c` by applying `f` to each element" intuitively makes...
A relevant comment -- modifying the behavior of `map` to allow for impure functions would make it much harder to change the behavior of `map` and `filter` in the future...
I'm honestly fine with pretty much any proposed behavior, but kicking the can down the road feels like the worst of both worlds. The longer we wait the more code...
Oh, I agree with that, but it should at least be clearly documented whether this is a bug or not. As you mentioned, the devs not tagging it as a...
It looks like [FillArrays](https://github.com/JuliaArrays/FillArrays.jl) has been operating under the assumption that the `map` documentation [implicitly allows](https://github.com/JuliaArrays/FillArrays.jl/issues/29) you to assume purity: > I believe the consensus is that the behaviour of...
Is anything holding this up? @dkarrasch @SebastianAment