how to specify type class in type of function?
[complete newbie using ghci 6.10.4 I understand that I should indicate that 'a' is in class Ord, since I use lessthan. But what is the syntax? Other style suggestions are welcome. mysort :: [a] -> [a] mysort [] = [] mysort [x] = [x] mysort [x, y] | (x <= y) = [x, y] | otherwise = [y, x] mysort l = (mysort s1) ++ (mysort s2) where (s1, s2) = splitAt (div (length l) 2) l Could not deduce (Ord a) from the context () arising from a use of `<=' at /home/gry/foo.hs:53:17-22 Possible fix: add (Ord a) to the context of the type signature for `mysort' In the expression: (x <= y) In a stmt of a pattern guard for the definition of `mysort': (x <= y) In the definition of `mysort': mysort [x, y] | (x <= y) = [x, y] | otherwise = [y, x]
The syntax is thus:
mysort :: (Ord a) => [a] -> [a]
On Thu, Jan 28, 2010 at 4:06 PM, george young
[complete newbie using ghci 6.10.4 I understand that I should indicate that 'a' is in class Ord, since I use lessthan. But what is the syntax? Other style suggestions are welcome.
mysort :: [a] -> [a] mysort [] = [] mysort [x] = [x] mysort [x, y] | (x <= y) = [x, y] | otherwise = [y, x] mysort l = (mysort s1) ++ (mysort s2) where (s1, s2) = splitAt (div (length l) 2) l
Could not deduce (Ord a) from the context () arising from a use of `<=' at /home/gry/foo.hs:53:17-22 Possible fix: add (Ord a) to the context of the type signature for `mysort' In the expression: (x <= y) In a stmt of a pattern guard for the definition of `mysort': (x <= y) In the definition of `mysort': mysort [x, y] | (x <= y) = [x, y] | otherwise = [y, x]
_______________________________________________ Beginners mailing list Beginners@haskell.org http://www.haskell.org/mailman/listinfo/beginners
Am Donnerstag 28 Januar 2010 22:06:36 schrieb george young:
[complete newbie using ghci 6.10.4 I understand that I should indicate that 'a' is in class Ord, since I use lessthan. But what is the syntax? Other style suggestions are welcome.
mysort :: [a] -> [a]
mysort :: Ord a => [a] -> [a] Generally, function :: (Class1 a, Class2 b) => a -> b -> c The part of the type signature before the '=>' is called the context, if it consits of only one constraint, the parentheses can be omitted.
mysort [] = [] mysort [x] = [x] mysort [x, y] | (x <= y) = [x, y]
| otherwise = [y, x]
mysort l = (mysort s1) ++ (mysort s2)
Doesn't work, mysort [3,2,1,4] ~> [2,3,1,4] you want merge here instead of (++) (how to write merge is left as an exercise).
where (s1, s2) = splitAt (div (length l) 2) l
If you have a long list, all those calls to length will take a loong time. Consider passing it to a helper function which takes the length of the list as a parameter.
Could not deduce (Ord a) from the context () arising from a use of `<=' at /home/gry/foo.hs:53:17-22 Possible fix: add (Ord a) to the context of the type signature for `mysort' In the expression: (x <= y) In a stmt of a pattern guard for the definition of `mysort': (x <= y) In the definition of `mysort': mysort [x, y]
| (x <= y) = [x, y] | otherwise = [y, x]
participants (3)
-
Daniel Fischer -
george young -
Jorden Mauro