Difference between revisions of "99 questions/Solutions/17"

From HaskellWiki
Jump to navigation Jump to search
 
m (added another fold solution)
(2 intermediate revisions by 2 users not shown)
Line 3: Line 3:
 
Do not use any predefined predicates.
 
Do not use any predefined predicates.
   
Solution using take and drop:
+
Solution using <hask>take</hask> and <hask>drop</hask>:
  +
 
<haskell>
 
<haskell>
 
split xs n = (take n xs, drop n xs)
 
split xs n = (take n xs, drop n xs)
 
</haskell>
 
</haskell>
   
  +
Or even simpler using <hask>splitAt</hask>:
Alternatively, we have the following recursive solution:
 
  +
  +
<haskell>
  +
split = flip splitAt
  +
</haskell>
  +
 
But these should clearly be considered "predefined predicates". Alternatively, we have the following recursive solution:
  +
 
<haskell>
 
<haskell>
 
split :: [a] -> Int -> ([a], [a])
 
split :: [a] -> Int -> ([a], [a])
Line 18: Line 26:
   
 
The same solution as above written more cleanly:
 
The same solution as above written more cleanly:
  +
 
<haskell>
 
<haskell>
 
split :: [a] -> Int -> ([a], [a])
 
split :: [a] -> Int -> ([a], [a])
Line 24: Line 33:
 
</haskell>
 
</haskell>
   
  +
A similar solution using foldl:
Note that this function, with the parameters in the other order, exists as <hask>splitAt</hask>.
 
  +
  +
<haskell>
  +
split :: [a] -> Int -> ([a], [a])
  +
split [] _ = ([], [])
  +
split list n
  +
| n < 0 = (list, [])
  +
| otherwise = (first output, second output)
  +
where output = foldl (\acc e -> if third acc > 0 then (first acc ++ [e], second acc, third acc - 1) else (first acc, second acc ++ [e], third acc)) ([], [], n) list
  +
</haskell>
  +
  +
Note that for the above code to work you must define your own first, second, and third functions for tuples containing three elements like so:
  +
  +
<haskell>
  +
first :: (a, b, c) -> a
  +
first (x, _, _) = x
  +
  +
second :: (a, b, c) -> b
  +
second (_, y, _) = y
  +
  +
third :: (a, b, c) -> c
  +
third (_, _, z) = z
  +
</haskell>
  +
  +
Another foldl solution without defining tuple extractors:
  +
<haskell>
  +
split :: [a] -> Int -> ([a],[a])
  +
split lst n = snd $ foldl helper (0,([],[])) lst
  +
where helper (i,(left,right)) x = if i >= n then (i+1,(left,right++[x])) else (i+1,(left++[x],right))
  +
</haskell>

Revision as of 00:31, 24 December 2011

(*) Split a list into two parts; the length of the first part is given.

Do not use any predefined predicates.

Solution using take and drop:

split xs n = (take n xs, drop n xs)

Or even simpler using splitAt:

split = flip splitAt

But these should clearly be considered "predefined predicates". Alternatively, we have the following recursive solution:

split :: [a] -> Int -> ([a], [a])
split []         _             = ([], [])
split l@(x : xs) n | n > 0     = (x : ys, zs)
                   | otherwise = ([], l)
    where (ys,zs) = split xs (n - 1)

The same solution as above written more cleanly:

split :: [a] -> Int -> ([a], [a])
split xs 0 = ([], xs)
split (x:xs) n = let (f,l) = split xs (n-1) in (x : f, l)

A similar solution using foldl:

split :: [a] -> Int -> ([a], [a])
split [] _ = ([], [])
split list n
  | n < 0 = (list, [])
  | otherwise  = (first output, second output)
    where output = foldl (\acc e -> if third acc > 0 then (first acc ++ [e], second acc, third acc - 1) else (first acc, second acc ++ [e], third acc)) ([], [], n) list

Note that for the above code to work you must define your own first, second, and third functions for tuples containing three elements like so:

first :: (a, b, c) -> a  
first (x, _, _) = x  
  
second :: (a, b, c) -> b  
second (_, y, _) = y  
  
third :: (a, b, c) -> c  
third (_, _, z) = z

Another foldl solution without defining tuple extractors:

split :: [a] -> Int -> ([a],[a])
split lst n = snd $ foldl helper (0,([],[])) lst
    where helper (i,(left,right)) x = if i >= n then (i+1,(left,right++[x])) else (i+1,(left++[x],right))