# 99 questions/Solutions/16

### From HaskellWiki

< 99 questions | Solutions(Difference between revisions)

Citizen428 (Talk | contribs) |
Citizen428 (Talk | contribs) m |
||

Line 70: | Line 70: | ||

dropEvery xs n = map fst $ filter (\(x,i) -> i `mod` n /= 0) $ zip xs [1..] | dropEvery xs n = map fst $ filter (\(x,i) -> i `mod` n /= 0) $ zip xs [1..] | ||

</haskell> | </haskell> | ||

− | The filter function can be simplified. | + | The filter function can be simplified as seen above: |

+ | <haskell> | ||

+ | dropEvery xs n = map fst $ filter ((n/=) . snd) $ zip xs [1..] | ||

+ | </haskell> |

## Revision as of 19:03, 6 December 2010

(**) Drop every N'th element from a list.

dropEvery :: [a] -> Int -> [a] dropEvery [] _ = [] dropEvery (x:xs) n = dropEvery' (x:xs) n 1 where dropEvery' (x:xs) n i = (if (n `divides` i) then [] else [x]) ++ (dropEvery' xs n (i+1)) dropEvery' [] _ _ = [] divides x y = y `mod` x == 0

An alternative iterative solution:

dropEvery :: [a] -> Int -> [a] dropEvery list count = helper list count count where helper [] _ _ = [] helper (x:xs) count 1 = helper xs count count helper (x:xs) count n = x : (helper xs count (n - 1))

Yet another iterative solution which divides lists using Prelude:

dropEvery :: [a] -> Int -> [a] dropEvery [] _ = [] dropEvery list count = (take (count-1) list) ++ dropEvery (drop count list) count

A similar approach using guards:

dropEvery :: [a] -> Int -> [a] dropEvery xs n | length xs < n = xs | otherwise = take (n-1) xs ++ dropEvery (drop n xs) n

Using zip:

dropEvery = flip $ \n -> map snd . filter ((n/=) . fst) . zip (cycle [1..n])

Using zip and list comprehensions

dropEvery :: [a] -> Int -> [a] dropEvery xs n = [ i | (i,c) <- ( zip xs [1,2..]), (mod c n) /= 0]

A more complicated approach which first divides the input list into sublists that do not contain the nth element, and then concatenates the sublists to a result list (if not apparent: the author's a novice):

dropEvery :: [a] -> Int -> [a] dropEvery [] _ = [] dropEvery xs n = concat (split n xs) where split _ [] = [] split n xs = fst splitted : split n ((safetail . snd) splitted) where splitted = splitAt (n-1) xs safetail xs | null xs = [] | otherwise = tail xs

First thing that came to mind:

dropEvery xs n = map fst $ filter (\(x,i) -> i `mod` n /= 0) $ zip xs [1..]

The filter function can be simplified as seen above:

dropEvery xs n = map fst $ filter ((n/=) . snd) $ zip xs [1..]