Hi I'm new in Haskell programming. I'm trying to implement operator "++" by myself. Here's a small program I wrote but it won't work:
append (es:e) xs =
if (null es)
then e:xs
else append es (e:xs)
I received lots of type error with [a],[[a]] and [[[a]]]. Still confusing about the list type in Haskell. Can someone help me with that? Thank you. :)
append (es:e) xs =
^^^^
Typically, you would write (e:es), which could be spoken "one e before a list of es". You actually have used this meaning below, but have told the compiler that es is an element and e is a list - which produces the type errors you received.
if (null es)
That's not how you should test for an emtpy list. In fact, if you'd call append [] … you'd get a "non-exhaustive pattern" error, because (e:es) is always a list of at least one element. So try two patterns:
append [] xs = xs
append (e:es) xs = append es (e:xs)
However, this still doesn't work expected - the first list is actually reversed by this snippet. Instead, the first element (e) needs to go before the whole rest of the list (es and xs):
append (e:es) xs = e : (append es xs)
(and that's indeed how ++ is implemented)
++ with xs ++ ys = augment (\c n -> foldr c n xs) ys in actual optimised GHC Haskell.
RULE thing meant :-)
augment gives optimization when it's combined with the fusion rule "foldr/augment" forall k z xs (g::forall b. (a->b->b) -> b -> b) . foldr k z (augment g xs) = g k (foldr k z xs), which ideally optimizes away most of the list allocation. augment differs from the similar build in that it's used to prepend stuff to a list rather than building the whole list. See GHC.Base.