大数据（7k）比较Scala、Java、SQL、Python的case模式匹配

0、关键词

match和case

1、匹配常量

Scala

for (i <- 0 until 3){
  val j = i match {
    case 0 => "零蛋"
    case 1 => "一血"
    case _ => "五杀"
  }
  println(i + j)

Java

for (int i = 0; i < 3; i++) {
    switch (i) {
        case 0:
            System.out.println(i + "零蛋");
            break;
        case 1:
            System.out.println(i + "一血");
            break;
        default:
            System.out.println(i + "五杀");
            break;
    }
}

Python

for i in range(3):
    j = {
        0: '零蛋',
        1: '一血',
    }.get(i, '五杀')
    print(str(i) + j)

打印结果

0零蛋
1一血
2五杀

sql

SELECT
    salary,
    (CASE
    WHEN salary>10000 THEN '(10k,+∞)'
    WHEN salary>5000 THEN '(5k,10k]'
    WHEN salary>3000 THEN '(3k,5k]'
    ELSE '(-∞,3k]'
    END) AS salary_section
FROM `employees`
;

2、匹配类型

def patternMatching(x: Any): String = x match {
  case i: Int => i + " 是整数"
  case i: String => i + " 是字符串"
  case i: List[_] => i + " 是列表"
  case i: Array[Int] => i.mkString("-") + " 数组"
  case _ => "其它"
}

println(patternMatching(-1))
println(patternMatching("a"))
println(patternMatching(List(2, "b")))
println(patternMatching(Array(3,2)))
println(patternMatching(Array("c","b")))

打印结果

-1 是整数
a 是字符串
List(2, b) 是列表
3-2 数组
其它

3、匹配数组

def patternMatching(x: Array[Int]): String = x match {
  case Array(10) => "精准匹配Array(10)"
  case Array(x) => "匹配含1个元素的数组" + x
  case Array(x, y) => "匹配含2个元素的数组" + x * y
  case Array(9, _*) => "匹配以9开头的数组"
  case a => a.mkString("-")
}

println(patternMatching(Array(10)))
println(patternMatching(Array(5)))
println(patternMatching(Array(7, 6)))
println(patternMatching(Array(9, 8 ,7,6)))
println(patternMatching(Array(10,9,8)))

打印结果

精准匹配Array(10)
匹配含1个元素的数组5
匹配含2个元素的数组42
匹配以9开头的数组
10-9-8

4、匹配元组

for (t <- Seq((0, 0), (5, 0), (5, 5), (0, 0, 1))) {
  val x = t match {
    case (0, b) => " 是第1个元素是0的两元元组 " + b
    case (a, 0) => " 是第2个元素是0的两元元组 " + a
    case (a:Int, b:Int) => " 是含2个元素的元组 " + (a * b)
    case _ => " 是其它"
  }
  println(t + x)
}

打印结果

(0,0) 是第1个元素是0的两元元组 0
(5,0) 是第2个元素是0的两元元组 5
(5,5) 是含2个元素的元组 25
(0,0,1) 是其它

4.1、for匹配元组

for (("a", x) <- Seq(("a", 1), ("b", 2), ("c", 3), ("a", 4))) {
  println(x)
}

打印结果

1
4

4.2、map匹配元组

val s: Seq[(String, Int)] = Seq(("a", 2), ("b", 3), ("a", 4))
val l: Seq[(String, Int)] = s.map {
  case ("b", c) => ("b", c * 100)
  case (w, c) => (w, c * 2)
}
println(l)

打印结果

List((a,4), (b,300), (a,8))

5、声明变量时使用匹配模式

5.1、一行代码 赋值多个

Scala

val (a, b) = ("A", 2)
println(a)
println(b)

Python

a, b = 'A', 2
print(a)
print(b)

5.2、一行代码 从列表中赋值多个

Scala

def f(ls:List[Int]):Unit ={
  val List(a, b, _*) = ls
  println(a, b)
}

f(List(1, 2, 3, 4))  // (1,2)
f(List(5, 6, 7, 8, 9))  // (5,6)

Python

def f(ls):
    a, b, *args = ls
    print(a, b, args)

f([1, 2, 3, 4])  # 1 2 [3, 4]
f([5, 6, 7, 8, 9])  # 5 6 [7, 8, 9]

6、for表达式中的模式匹配

val map = Map("A" -> 1, "B" -> 0, "C" -> 3)

//全部k-v
for ((k, v) <- map) {
  println(k + v)
}

//遍历value=0的k-v：B
for ((k, 0) <- map) {
  println(k + " -> " + 0) //打印结果：B
}

//遍历v>=1的k-v：A和C
for ((k, v) <- map if v >= 1) {
  println(k + " --> " + v)
}

7、偏函数中的模式匹配（工作常用）

• 英文名：partial function
• 简述：不完整的函数；函数的某个部分

//将列表中的Int类型的元素加1，并去掉字符串
val ls1 = List(1,2,3,4,5,"a")
val ls2 = ls1.collect { case x: Int => x + 1 }