Github user chenghao-intel commented on a diff in the pull request:
https://github.com/apache/spark/pull/7886#discussion_r36500358
--- Diff:
sql/core/src/test/scala/org/apache/spark/sql/execution/PlannerSuite.scala ---
@@ -170,14 +170,57 @@ class PlannerSuite extends SparkFunSuite with
SQLTestUtils {
// Disable broadcast join
withSQLConf(SQLConf.AUTO_BROADCASTJOIN_THRESHOLD.key -> "-1") {
+ val joins = Array("JOIN", "LEFT OUTER JOIN", "RIGHT OUTER JOIN",
"FULL OUTER JOIN")
+ var i = 0
+ while (i < joins.length) {
+ var j = 0
+ while (j < joins.length) {
+ val firstJoin: String = joins(i)
+ val secondJoin: String = joins(j)
+
+ {
+ val numExchanges: Int = sql(
+ s"""
+ |SELECT *
+ |FROM
+ | normal $firstJoin small ON (normal.key = small.key)
+ | $secondJoin tiny ON (small.key = tiny.key)
+ """.stripMargin
+ ).queryExecution.executedPlan.collect {
+ case exchange: Exchange => exchange
+ }.length
+ assert(numExchanges === 3)
+ }
+
+ {
+ val numExchanges: Int = sql(
+ s"""
+ |SELECT *
+ |FROM
+ | normal $firstJoin small ON (normal.key = small.key)
+ | $secondJoin tiny ON (normal.key = tiny.key)
+ """.stripMargin
+ ).queryExecution.executedPlan.collect {
+ case exchange: Exchange => exchange
+ }.length
+ assert(numExchanges === 3)
+ }
+
+ j += 1
+ }
+ i += 1
+ }
+
{
- val numExchanges = sql(
- """
- |SELECT *
- |FROM
- | normal JOIN small ON (normal.key = small.key)
- | JOIN tiny ON (small.key = tiny.key)
- """.stripMargin
+ val numExchanges: Int = sql(
+ s"""
+ |SELECT small.key, count(*)
+ |FROM
+ | normal JOIN small ON (normal.key = small.key)
+ | JOIN tiny ON (small.key = tiny.key)
+ |GROUP BY
+ | small.key
+ """.stripMargin
--- End diff --
Also add unit test like:
```
SELECT small.key, count(*)
FROM
normal JOIN small ON (normal.key = small.key)
LEFT JOIN tiny ON (small.key = tiny.key)
GROUP BY
small.key
```
Since the left outer join will not create new null keys, we expect the
aggregation does not require any shuffle from join operator.
---
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]
