N.B. I haven't thought this through carefully.
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Extension methods have subtly different semantics than direct members though, and they're *dramatically* less discoverable with tooling (scaladoc or IDE/Metals). I do agree that the magic right associativity is weird, but is it actually made better by generalizing it?
-
You don't just have to think about definition site here. Think about the call-site. Right now, we know things associate right when they end with :. What happens if *anything* can be right associative, and the only way to know is to look at the definition?
- Show replies
New conversation -
-
-
Just do away with right-associative operators altogether! The only use-cases are: - Lists (which are a bad data-structure) - prepend That's way too little benefit for the complexity cost involved. I have done away with them (along with unary operators) and it's totally fine.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
(Considering Scala's design "trajectory", v3 will probably "vastly simplify" the whole matter by "generalizing" left-/right-associativity to also include up-/down-associativity, while adding a new keyword that allows freely defining associativity in 360 1° steps.)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
APL has right association up the wazoo
-
although reading the whole thread this might not be useful information ;-) soz
- Show replies
New conversation -
-
-
I did end up using it for decomposition very often, but not very useful otherwise.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.