Github user dongjoon-hyun commented on a diff in the pull request:
https://github.com/apache/spark/pull/14048#discussion_r69662351
--- Diff: sql/core/src/main/scala/org/apache/spark/sql/Dataset.scala ---
@@ -184,7 +183,8 @@ class Dataset[T] private[sql](
// to happen right away to let these side effects take place eagerly.
case p if hasSideEffects(p) =>
LogicalRDD(queryExecution.analyzed.output,
queryExecution.toRdd)(sparkSession)
- case Union(children) if children.forall(hasSideEffects) =>
+ case Union(children) if children.forall(x =>
+ x.isInstanceOf[InsertIntoTable] ||
x.isInstanceOf[InsertIntoHadoopFsRelationCommand]) =>
--- End diff --
Please consider the following example in the current master. This PR blocks
this kind of weird behavior.
```
scala> spark.emptyDataFrame.createOrReplaceTempView("empty")
scala> sql("CREATE TABLE tbl(i INT) USING PARQUET")
scala> sql("(INSERT INTO tbl VALUES(1)) UNION ALL (INSERT INTO tbl
VALUES(2))")
scala> sql("select * from tbl").collect()
res4: Array[org.apache.spark.sql.Row] = Array([1], [2])
scala> sql("(INSERT INTO tbl VALUES(5)) UNION ALL (SELECT * FROM empty)")
scala> sql("select * from tbl").collect()
res6: Array[org.apache.spark.sql.Row] = Array([1], [2])
```
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
