[SPARK-17981][SPARK-17957][SQL] Fix Incorrect Nullability Setting to False in FilterExec

### What changes were proposed in this pull request?

When `FilterExec` contains `isNotNull`, which could be inferred and pushed down or users specified, we convert the nullability of the involved columns if the top-layer expression is null-intolerant. However, this is not correct, if the top-layer expression is not a leaf expression, it could still tolerate the null when it has null-tolerant child expressions.

For example, `cast(coalesce(a#5, a#15) as double)`. Although `cast` is a null-intolerant expression, but obviously`coalesce` is null-tolerant. Thus, it could eat null.

When the nullability is wrong, we could generate incorrect results in different cases. For example,

``` Scala
    val df1 = Seq((1, 2), (2, 3)).toDF("a", "b")
    val df2 = Seq((2, 5), (3, 4)).toDF("a", "c")
    val joinedDf = df1.join(df2, Seq("a"), "outer").na.fill(0)
    val df3 = Seq((3, 1)).toDF("a", "d")
    joinedDf.join(df3, "a").show
```

The optimized plan is like

```
Project [a#29, b#30, c#31, d#42]
+- Join Inner, (a#29 = a#41)
   :- Project [cast(coalesce(cast(coalesce(a#5, a#15) as double), 0.0) as int) AS a#29, cast(coalesce(cast(b#6 as double), 0.0) as int) AS b#30, cast(coalesce(cast(c#16 as double), 0.0) as int) AS c#31]
   :  +- Filter isnotnull(cast(coalesce(cast(coalesce(a#5, a#15) as double), 0.0) as int))
   :     +- Join FullOuter, (a#5 = a#15)
   :        :- LocalRelation [a#5, b#6]
   :        +- LocalRelation [a#15, c#16]
   +- LocalRelation [a#41, d#42]
```

Without the fix, it returns an empty result. With the fix, it can return a correct answer:

```
+---+---+---+---+
|  a|  b|  c|  d|
+---+---+---+---+
|  3|  0|  4|  1|
+---+---+---+---+
```
### How was this patch tested?

Added test cases to verify the nullability changes in FilterExec. Also added a test case for verifying the reported incorrect result.

Author: gatorsmile <[email protected]>

Closes #15523 from gatorsmile/nullabilityFilterExec.

Author: gatorsmile

sql/core/src/main/scala/org/apache/spark/sql/execution/basicPhysicalOperators.scala

@@ -90,7 +90,13 @@ case class FilterExec(condition: Expression, child: SparkPlan)
 
   // Split out all the IsNotNulls from condition.
   private val (notNullPreds, otherPreds) = splitConjunctivePredicates(condition).partition {
-    case IsNotNull(a: NullIntolerant) if a.references.subsetOf(child.outputSet) => true
+    case IsNotNull(a) => isNullIntolerant(a) && a.references.subsetOf(child.outputSet)
+    case _ => false
+  }
+
+  // If one expression and its children are null intolerant, it is null intolerant.
+  private def isNullIntolerant(expr: Expression): Boolean = expr match {
+    case e: NullIntolerant => e.children.forall(isNullIntolerant)
     case _ => false
   }
 

sql/core/src/test/scala/org/apache/spark/sql/DataFrameSuite.scala

@@ -28,8 +28,8 @@ import org.scalatest.Matchers._
 
 import org.apache.spark.SparkException
 import org.apache.spark.sql.catalyst.TableIdentifier
-import org.apache.spark.sql.catalyst.plans.logical.{OneRowRelation, Project, Union}
-import org.apache.spark.sql.execution.QueryExecution
+import org.apache.spark.sql.catalyst.plans.logical.{Filter, OneRowRelation, Project, Union}
+import org.apache.spark.sql.execution.{FilterExec, QueryExecution}
 import org.apache.spark.sql.execution.aggregate.HashAggregateExec
 import org.apache.spark.sql.execution.exchange.{BroadcastExchangeExec, ReusedExchangeExec, ShuffleExchange}
 import org.apache.spark.sql.functions._
@@ -1635,6 +1635,76 @@ class DataFrameSuite extends QueryTest with SharedSQLContext {
     }
   }
 
+  private def verifyNullabilityInFilterExec(
+      df: DataFrame,
+      expr: String,
+      expectedNonNullableColumns: Seq[String]): Unit = {
+    val dfWithFilter = df.where(s"isnotnull($expr)").selectExpr(expr)
+    // In the logical plan, all the output columns of input dataframe are nullable
+    dfWithFilter.queryExecution.optimizedPlan.collect {
+      case e: Filter => assert(e.output.forall(_.nullable))
+    }
+
+    dfWithFilter.queryExecution.executedPlan.collect {
+      // When the child expression in isnotnull is null-intolerant (i.e. any null input will
+      // result in null output), the involved columns are converted to not nullable;
+      // otherwise, no change should be made.
+      case e: FilterExec =>
+        assert(e.output.forall { o =>
+          if (expectedNonNullableColumns.contains(o.name)) !o.nullable else o.nullable
+        })
+    }
+  }
+
+  test("SPARK-17957: no change on nullability in FilterExec output") {
+    val df = sparkContext.parallelize(Seq(
+      null.asInstanceOf[java.lang.Integer] -> new java.lang.Integer(3),
+      new java.lang.Integer(1) -> null.asInstanceOf[java.lang.Integer],
+      new java.lang.Integer(2) -> new java.lang.Integer(4))).toDF()
+
+    verifyNullabilityInFilterExec(df,
+      expr = "Rand()", expectedNonNullableColumns = Seq.empty[String])
+    verifyNullabilityInFilterExec(df,
+      expr = "coalesce(_1, _2)", expectedNonNullableColumns = Seq.empty[String])
+    verifyNullabilityInFilterExec(df,
+      expr = "coalesce(_1, 0) + Rand()", expectedNonNullableColumns = Seq.empty[String])
+    verifyNullabilityInFilterExec(df,
+      expr = "cast(coalesce(cast(coalesce(_1, _2) as double), 0.0) as int)",
+      expectedNonNullableColumns = Seq.empty[String])
+  }
+
+  test("SPARK-17957: set nullability to false in FilterExec output") {
+    val df = sparkContext.parallelize(Seq(
+      null.asInstanceOf[java.lang.Integer] -> new java.lang.Integer(3),
+      new java.lang.Integer(1) -> null.asInstanceOf[java.lang.Integer],
+      new java.lang.Integer(2) -> new java.lang.Integer(4))).toDF()
+
+    verifyNullabilityInFilterExec(df,
+      expr = "_1 + _2 * 3", expectedNonNullableColumns = Seq("_1", "_2"))
+    verifyNullabilityInFilterExec(df,
+      expr = "_1 + _2", expectedNonNullableColumns = Seq("_1", "_2"))
+    verifyNullabilityInFilterExec(df,
+      expr = "_1", expectedNonNullableColumns = Seq("_1"))
+    // `constructIsNotNullConstraints` infers the IsNotNull(_2) from IsNotNull(_2 + Rand())
+    // Thus, we are able to set nullability of _2 to false.
+    // If IsNotNull(_2) is not given from `constructIsNotNullConstraints`, the impl of
+    // isNullIntolerant in `FilterExec` needs an update for more advanced inference.
+    verifyNullabilityInFilterExec(df,
+      expr = "_2 + Rand()", expectedNonNullableColumns = Seq("_2"))
+    verifyNullabilityInFilterExec(df,
+      expr = "_2 * 3 + coalesce(_1, 0)", expectedNonNullableColumns = Seq("_2"))
+    verifyNullabilityInFilterExec(df,
+      expr = "cast((_1 + _2) as boolean)", expectedNonNullableColumns = Seq("_1", "_2"))
+  }
+
+  test("SPARK-17957: outer join + na.fill") {
+    val df1 = Seq((1, 2), (2, 3)).toDF("a", "b")
+    val df2 = Seq((2, 5), (3, 4)).toDF("a", "c")
+    val joinedDf = df1.join(df2, Seq("a"), "outer").na.fill(0)
+    val df3 = Seq((3, 1)).toDF("a", "d")
+    checkAnswer(joinedDf.join(df3, "a"), Row(3, 0, 4, 1))
+  }
+
   test("SPARK-17123: Performing set operations that combine non-scala native types") {
     val dates = Seq(
       (new Date(0), BigDecimal.valueOf(1), new Timestamp(2)),