Out of curiosity, what is the rationale for allowing programs to infer certain types, but not for the (inferred) type to be declared?



On Fri, Feb 15, 2013 at 4:54 PM, David McBride <toad3k@gmail.com> wrote:
sum' [] = []  -- returns a list of something the way you intended
sum' (x:xs) = x + xum' xs -- you intended it not to return a list but it could if you think about it.

The compiler says I think returns a list based on what I see so far, well if you can add these together then the only way you could get a list from that is if you were adding two lists together ie (+ l1 l2) :: [a] -> [a] -> [a].  That works if we assume that sum' must have accepted [[a]] and returned [a].

But in order for that to be the case [a] must be an instance of Num, otherwise they couldn't be added together like that, so tack on a Num [a] requirement on.

But having a typeclass of the form [a] that requires an extension, FlexibleContexts, which you can read about here: http://www.haskell.org/ghc/docs/7.0.2/html/users_guide/other-type-extensions.html#flexible-contexts

On Fri, Feb 15, 2013 at 2:33 AM, sheng chen <kkchensheng@gmail.com> wrote:
Hi,

I was puzzled by the following little program.

sum' [] = []
sum' (x:xs) = x + sum' xs

I thought the GHC type checker will report a type error. However, the type checker accepts this program and gives the type

Num [a] => [[a]] -> [a]

When I add type annotation to the program

sum' :: Num [a] => [[a]] -> [a]
sum' [] = []
sum' (x:xs) = x + sum' xs

The GHC asks me to add FlexibleContexts language extension.

I would appreciate explanation on this issue.

Thanks,
Sheng

_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe



_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe