自定义数据类型过滤器 - 任何数字类型

Question

import System.IO()

data Point = Point
    { pointX :: {-# UNPACK #-} !Double  -- ^ X coordinate
    , pointY :: {-# UNPACK #-} !Double  -- ^ Y coordinate
    } deriving (Show, Eq)

data Polygon = Polygon
    { points :: [Point]
    , yvalue :: Int
    } deriving (Show)

1. 创建文件test.hs
2. 将上面的两个自定义数据类型定义复制进去
3. 打开ghci并输入:l test.hs

4. 输入这个测试用例：

   *test> let a = Polygon {points = [Point {pointX = 0.0, pointY = 0.0},Point {pointX = 4.0, pointY = 0.0},Point {pointX = 4.0, pointY = 2.0},Point {pointX = 4.0, pointY = 4.0},Point {pointX = 0.0, pointY = 4.0},Point {pointX = 0.0, pointY = 2.0},Point {pointX = 0.0, pointY = 0.0}], yvalue = 2}
   

5. 然后：

   *test> let x = filter (<=(fromIntegral (yvalue a)).pointY) $ points a  
   

您收到以下错误：

* couldn't match type `Point` with `Point -> c'
  Expected type: [Point -> c]
    Actual type: [Point]
* In the second argument of `($)', namely `points a'
  In the expression:
    filter (<=(fromIntegral (yvalue a)) . pointY) $ points a
  In an equation for `x':
    x = filter (<=(fromIntegral (yvalue a)).pointY) $ points a  
* Relevant bindings include
    x :: [Point -> c] (bound at <interactive>:92:5)

Answer 1

解决办法是：

如果您希望按特定的数字数据类型字段进行过滤，那么您只需实现一个比较实例。

就我而言：

instance Ord Point where
    compare x y = compare (pointY x) (pointY y)

这将我希望比较的指定字段设置为 pointY，它已经是 Double 的实例。

然后我可以使用以下过滤器：

filter(\x -> x > Point {pointX = 0.0, pointY = (fromIntegral(yvalue))} $ points a

注意：这个过滤器中的 pointX 是完全不相关的。

可能会出现另一种解决方案，因为这里有一些出色的 Haskell 程序员，但这个可行。