Hello,
My friend reached the following version:
chop :: String -> String
chop = foldr go []
where
go x xs
| isSpace x && null xs = []
| otherwise = x:xs
This version is faster than the reverse version in most cases. The
point is checking "isSpace" first and falling into "otherwise" in many
cases, which is a natural co-recursion.
Thanks anyway.
--Kazu
>> Hello,
>>
>> Of course, I use ByteString or Text for real programming. But I would
>> like to know whether or not there are any efficient methods to remove
>> a tail part of a list.
>>
>> --Kazu
>
> In that case, I would prefer this version, since it is lazier:
>
> lazyChop :: String -> String
> lazyChop s = pref ++ if null s' then [] else (mid_sp ++ lazyChop s')
> where
> (pref,sp_suf) = break isSpace s
> (mid_sp,s') = span isSpace sp_suf
>
> By "lazier" I mean:
>
> *Main> chopReverse $ "hello world " ++ undefined
> "*** Exception: Prelude.undefined
> *Main> chopFoldr $ "hello world " ++ undefined
> "*** Exception: Prelude.undefined
> *Main> lazyChop $ "hello world " ++ undefined
> "hello world*** Exception: Prelude.undefined
>
> Daniel
_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
