kosmikus, 28 days ago

@DiazCarrete I think both are nice to have.

BoydStephenSmithJr, 28 days ago

@DiazCarrete Ooh, haven't read the article, but despite my distain for TypeApplications, I'm already excited by the name of the new extension.

BoydStephenSmithJr, 28 days ago (edited 28 days ago)

@DiazCarrete If there's ever a new report NoListTuplePuns needs to be included. Also, more developers should unpun their newtypes and records.

I still prefer Proxy-ed APIs for compatibility, but it will be nice to be able to use TypeAbstractions for my higher rank lamdas / local bindings. Definitely an improvement over scoped type variables, and I don't think it even stole that much pattern syntax; as-patterns were almost always written with no whitespace.

DiazCarrete, 28 days ago

@BoydStephenSmithJr How do you un-pun your newtypes? Do you use some kind of "Mk-" prefix?

BoydStephenSmithJr, 28 days ago

@DiazCarrete That's what I do, yes. newtype Semantic2 = MkSemantic2 { unSemantic2 :: Semantic1 }

Though sometimes I do something different, like newtype Subset = AssertSubset { asSuperset :: Superset } for things where the rep. sharing is more practical, even if the syntax isn't actually the same, like Email/Text or PhoneNumber/Int64 since construction will generally be through something like isEmail :: Text -> Either ParseFailure Email or inputPhone :: Natural -> Maybe PhoneNumber