scala

问题 I have a Spark DataFrame with the following structure: root |-- distribution: vector (nullable = true) +--------------------+ | topicDistribution| +--------------------+ | [0.1, 0.2] | | [0.3, 0.2] | | [0.5, 0.2] | | [0.1, 0.7] | | [0.1, 0.8] | | [0.1, 0.9] | +--------------------+ My question is: How to add a column with the index of the maximum value for each row? It should be something like this: root |-- distribution: vector (nullable = true) |-- max_index: integer (nullable = true) +----

问题 The sbt documentation shows example of how to declare dependency only between projects. But I'm positive that there are ways to declare one config be dependent on another, just like the Test configuration uses classpath from the Compile configuration. How can I declare my own configuration so that it would depend on the Compile config generated classpath? I take a more close look to suggested solution and bunch of question arose again. So I reopen the question I can not deduce sbt behavior

问题 The sbt documentation shows example of how to declare dependency only between projects. But I'm positive that there are ways to declare one config be dependent on another, just like the Test configuration uses classpath from the Compile configuration. How can I declare my own configuration so that it would depend on the Compile config generated classpath? I take a more close look to suggested solution and bunch of question arose again. So I reopen the question I can not deduce sbt behavior

问题 I have a list: List(1,2,3,4,5,6) that I would like to to convert to the following map: Map(1->2,3->4,5->6) How can this be done? 回答1: Mostly resembles @Vakh answer, but with a nicer syntax: val l = List(1,2,3,4,5,6) val m = l.grouped(2).map { case List(key, value) => key -> value}.toMap // Map(1 -> 2, 3 -> 4, 5 -> 6) 回答2: Try: val l = List(1,2,3,4,5,6) val m = l.grouped(2).map(l => (l(0), l(1))).toMap 回答3: if the list is guaranteed to be of even length: val l = List(1,2,3,4,5,6) val m = l

问题 I have a list: List(1,2,3,4,5,6) that I would like to to convert to the following map: Map(1->2,3->4,5->6) How can this be done? 回答1: Mostly resembles @Vakh answer, but with a nicer syntax: val l = List(1,2,3,4,5,6) val m = l.grouped(2).map { case List(key, value) => key -> value}.toMap // Map(1 -> 2, 3 -> 4, 5 -> 6) 回答2: Try: val l = List(1,2,3,4,5,6) val m = l.grouped(2).map(l => (l(0), l(1))).toMap 回答3: if the list is guaranteed to be of even length: val l = List(1,2,3,4,5,6) val m = l

问题 I have a list: List(1,2,3,4,5,6) that I would like to to convert to the following map: Map(1->2,3->4,5->6) How can this be done? 回答1: Mostly resembles @Vakh answer, but with a nicer syntax: val l = List(1,2,3,4,5,6) val m = l.grouped(2).map { case List(key, value) => key -> value}.toMap // Map(1 -> 2, 3 -> 4, 5 -> 6) 回答2: Try: val l = List(1,2,3,4,5,6) val m = l.grouped(2).map(l => (l(0), l(1))).toMap 回答3: if the list is guaranteed to be of even length: val l = List(1,2,3,4,5,6) val m = l

问题 This code fails to compile: object Foo { implicit def toString(i: Int): String = i.toString def foo(x: String) = println(x) foo(23) } Above code fails to compile with following error: error: type mismatch; found : scala.this.Int(23) required: String foo(23) But, this code compiles object Foo { implicit def asString(i: Int): String = i.toString def foo(x: String) = println(x) foo(23) } Why does the name of an implicit def should matter? Note: If the method is named equals , it also does not

问题 This code fails to compile: object Foo { implicit def toString(i: Int): String = i.toString def foo(x: String) = println(x) foo(23) } Above code fails to compile with following error: error: type mismatch; found : scala.this.Int(23) required: String foo(23) But, this code compiles object Foo { implicit def asString(i: Int): String = i.toString def foo(x: String) = println(x) foo(23) } Why does the name of an implicit def should matter? Note: If the method is named equals , it also does not

问题 I'm trying to test a class Cls with two functions: A and B . A loads a DataFrame and B calls A then does some operations and returns a new DataFrame . For the sake of example: class Cls { def A(dummy: Int): Int = 5 def B(): Int = A(7) + 1 } With Scalamock how can write my test code ? I tried: test("test case") { val f = stub[Cls] f.A _ when 7 returns 5 assert(f.B() == 6) } I expect test passed successfully and I get 0 did not equal 6 (mytestcase.scala:24) (I do understand that that scalamock

问题 I feel a little stupid for asking this, but I'm currently learning functional programming and completed an exercise on creating singly linked lists and it just got me thinking, is it even possible to create an immutable doubly linked list? Suppose the list A :: B, at time of construction, A needs to know about B, but B also needs to know about A. I've been doing this in Scala, so I'm not sure if it's specific to Scala, but I can't picture how that would work. I'm not looking for a substitute