David Gold

I worked on developing the NullableArrays package. I'm happy to help navigate performance issues, if I can. If there are requests for features, please do file an issue.

The package currently is more or less at feature parity with DataArrays, modulo the PooledDataArray interface/functionality of the latter. That functionality is to be refactored into a different package that...

@johnmyleswhite I do as well, but it's kind of unfortunate that one has to include an entire copy of a function's code just to tweak the conditionals and nothing else....

Thanks, Milan. I do have something to add, which I'll do shortly.

Also, maybe NullableArrays 0.1 could serve as a deprecation period for Julia 0.4 support?

@nalimilan IMO, `NullableArray`s are _mostly_ efficient replacement for `Array{Nullable}` and should _mostly_ have identical behavior. The exceptions I'm thinking of are things like `map`, `reduce`, `mean`, etc. in which I...

> Yes, but then were does this list end? It's not great to tell users that they have to learn a (possibly evolving) list of functions which don't behave normally....

> I'm really not a fan of a design where the semantics of a function change depending on the call context. In the case of `map(f, X::NullableArray)`, the semantics of...

@davidanthoff Can you help me to understand your concern? I'm proposing "automatic lifting" in the context of higher-order functions such as `map`. So, if `f(x::T)` is an extant method, `map(f,...

I don't know. My current solution would be to `@collect i in X |> select(i & false)`, since the macro context allows us to do syntax manipulations that would result...