Daniel Winograd-Cort
Daniel Winograd-Cort
> I gave this a try and it seems to work fine for most cases but for very large matrices the compiler simplifier exhausts the ticks. Changing to this type...
It would be great if this were implemented. There is https://ghc.haskell.org/trac/ghc/ticket/13116, but no work has been done on it.
This is very cool, and thanks so much for upstreaming! I have a busy weekend, so it might take me a few days to get to this, but I definitely...
If I'm reading your code correctly, the basic idea is that you create a record containing a `Dict c t` for every `t` in the row-type, and then you index...
@gnumonik Hey, I just wanted to apologize for taking so long to respond. I haven't forgotten about you, but I just have a bunch of other stuff on my plate...
I'm not exactly sure how partial application would work. Is this just a short-hand for writing the type signature? Like, are you imagining a type family that you write like...
Yeah, ok, we can do something like Caml's named arguments. Consider this function: ```haskell infixl 3 .$ (.$) :: forall l t r r' x. (KnownSymbol l, r' .! l...
I thought about `Optional` a little more, and the fundamental issue is that there's no easy way to do type equality at the value level without a `Typeable` constraint, which...
The main reason I've been struggling with `Optional` is the scenario where you have a record like: ```haskell rec :: Rec ("x" .== Integer .+ "y" .== String) rec =...
Now that I went to code it up, I totally agree that option 2 makes the most sense: ```haskell class Optional (l :: Symbol) (a :: k) (r :: Row...