ueshin commented on a change in pull request #24232: [SPARK-27297] [SQL] Add 
higher order functions to scala API
URL: https://github.com/apache/spark/pull/24232#discussion_r316376371
 
 

 ##########
 File path: 
sql/core/src/test/scala/org/apache/spark/sql/DataFrameFunctionsSuite.scala
 ##########
 @@ -1917,19 +1921,33 @@ class DataFrameFunctionsSuite extends QueryTest with 
SharedSparkSession {
       null
     ).toDF("i")
 
+    // transform(i, x -> x + 1)
+    val resA = Seq(
+      Row(Seq(2, 10, 9, 8)),
+      Row(Seq(6, 9, 10, 8, 3)),
+      Row(Seq.empty),
+      Row(null))
+
+    // transform(i, (x, i) -> x + i)
+    val resB = Seq(
+      Row(Seq(1, 10, 10, 10)),
+      Row(Seq(5, 9, 11, 10, 6)),
+      Row(Seq.empty),
+      Row(null))
+
     def testArrayOfPrimitiveTypeNotContainsNull(): Unit = {
-      checkAnswer(df.selectExpr("transform(i, x -> x + 1)"),
-        Seq(
-          Row(Seq(2, 10, 9, 8)),
-          Row(Seq(6, 9, 10, 8, 3)),
-          Row(Seq.empty),
-          Row(null)))
-      checkAnswer(df.selectExpr("transform(i, (x, i) -> x + i)"),
-        Seq(
-          Row(Seq(1, 10, 10, 10)),
-          Row(Seq(5, 9, 11, 10, 6)),
-          Row(Seq.empty),
-          Row(null)))
+      checkAnswer(df.selectExpr("transform(i, x -> x + 1)"), resA)
+      checkAnswer(df.selectExpr("transform(i, (x, i) -> x + i)"), resB)
+
+      checkAnswer(df.select(transform(col("i"), x => x + 1)), resA)
+      checkAnswer(df.select(transform(col("i"), (x, i) => x + i)), resB)
+
+      checkAnswer(df.select(transform(col("i"), new JFunc {
+        def call(x: Column) = x + 1
+      })), resA)
+      checkAnswer(df.select(transform(col("i"), new JFunc2 {
+        def call(x: Column, i: Column) = x + i
+      })), resB)
 
 Review comment:
   IIUC:
   1. The Java compiler tries to resolve lambdas as a Single Abstract Method 
(SAM) type.
   2. The SAM type has only one abstract method and can have other non-abstract 
methods including default methods for interfaces. Not only interfaces but also 
abstract classes can be a SAM type.
   3. `@FunctionalInterface` is just a marker for Java compiler which implies 
the annotated interface or class is the SAM type. SAM types are not necessarily 
annotated with `@FunctionalInterface`.
   4. The scala `Function` satisfies the SAM type requirements:
      -  Scala's trait is basically an interface.
      - `Function` has only one `apply()` abstract method.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to