简体   繁体   English

Haskell。 按元组列表中的第二个元素排序

[英]Haskell. Sort by second element in the list of tuples

I have a code histogram , which shows the number of all characters in String.我有一个代码histogram ,它显示了字符串中所有字符的数量。

histogram :: String -> [(Char, Int)]
histogram = foldr help [] where
    help x [] = [(x,1)] 
    help x ((y,z):yzs)
      | x == y = (y,z+1) : yzs 
      | otherwise = (y,z) : help x yzs

But the result is shown in a chaotic order.但结果以混乱的顺序显示。 I want to sort the result by the second element of the list (ie by numbers).我想按列表的第二个元素(即按数字)对结果进行排序。 For this I implemented ssort code but it sorts only by the first element (char).为此,我实现ssort代码,但它仅按第一个元素(char)排序。

import Data.List (minimum, delete)

ssort :: Ord t => [t] -> [t]
ssort [] = []
ssort xs = x : ssort (remove x xs) where
     x = minimum xs

remove :: Eq a => a -> [a] -> [a]
remove _ [] = [] 
remove y (x:xs) | y == x = xs 
    | otherwise = x : remove y xs
 

Input: ssort [('a',1), ('g', 6), ('o',2), ('f',0)]输入: ssort [('a',1), ('g', 6), ('o',2), ('f',0)]

Output: [('a',1),('f',0),('g',6),('o',2)] (instead of [('f',0),('a',1),('o',2),('g',6)]) Output: [('a',1),('f',0),('g',6),('o',2)] (而不是 [('f',0),('a ',1),('o',2),('g',6)])

I also don't want to use any other builtins, as sort, sortOn and sortBy, I want to implement my own function.我也不想使用任何其他内置函数,如 sort、sortOn 和 sortBy,我想实现自己的 function。

How can I make it so that only the second elements are compared?我怎样才能使它只比较第二个元素?

You may provide your own definition of minimumBy :您可以提供自己的minimumBy定义:

minimumBy2 :: Ord t2 => [(t1, t2)] -> (t1, t2)
minimumBy2 (h:t) = go h t where
  go x [] = x
  go x (h0:t0) = go (min x h0) t0
minimumBy2 [] = error "minimumBy2: Empty list"

You can then use it in place of minimum in ssort and it should work.然后,您可以在ssort中使用它代替minimum ,它应该可以工作。

Assuming you don't want to use minimumBy:: Foldable t => (a -> a -> Ordering) -> ta -> a as well.假设您也不想使用minimumBy:: Foldable t => (a -> a -> Ordering) -> ta -> a

You can take advantage on how are tuples compared – first by the first element and then by the second.您可以利用元组的比较方式——首先是第一个元素,然后是第二个元素。 Therefore you can first swap all of the elements of your histogram, then sort and finally swap back:因此,您可以首先交换直方图的所有元素,然后排序并最后交换回来:

ssortOn2 :: (Ord t1, Ord t2) => [(t1, t2)] -> [(t1, t2)]
ssortOn2 = map swap . ssort . map swap 

(leaving all other functions unchanged) (保持所有其他功能不变)

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM