Github user kiszk commented on a diff in the pull request:
https://github.com/apache/spark/pull/19803#discussion_r152899229
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/WholeStageCodegenExec.scala
---
@@ -108,20 +108,22 @@ trait CodegenSupport extends SparkPlan {
/**
* Consume the generated columns or row from current SparkPlan, call its
parent's `doConsume()`.
+ *
+ * Note that `outputVars` and `row` can't both be null.
*/
final def consume(ctx: CodegenContext, outputVars: Seq[ExprCode], row:
String = null): String = {
val inputVars =
- if (row != null) {
+ if (outputVars != null) {
+ assert(outputVars.length == output.length)
--- End diff --
Is it better to add `assert(row == null)` to ensure your comment below?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
