Github user davies commented on a diff in the pull request:
https://github.com/apache/spark/pull/11210#discussion_r53684255
--- Diff: sql/core/src/test/scala/org/apache/spark/sql/JoinSuite.scala ---
@@ -63,36 +63,40 @@ class JoinSuite extends QueryTest with SharedSQLContext
{
test("join operator selection") {
sqlContext.cacheManager.clearCache()
- Seq(
- ("SELECT * FROM testData LEFT SEMI JOIN testData2 ON key = a",
classOf[LeftSemiJoinHash]),
- ("SELECT * FROM testData LEFT SEMI JOIN testData2",
classOf[LeftSemiJoinBNL]),
- ("SELECT * FROM testData JOIN testData2", classOf[CartesianProduct]),
- ("SELECT * FROM testData JOIN testData2 WHERE key = 2",
classOf[CartesianProduct]),
- ("SELECT * FROM testData LEFT JOIN testData2",
classOf[CartesianProduct]),
- ("SELECT * FROM testData RIGHT JOIN testData2",
classOf[CartesianProduct]),
- ("SELECT * FROM testData FULL OUTER JOIN testData2",
classOf[CartesianProduct]),
- ("SELECT * FROM testData LEFT JOIN testData2 WHERE key = 2",
classOf[CartesianProduct]),
- ("SELECT * FROM testData RIGHT JOIN testData2 WHERE key = 2",
classOf[CartesianProduct]),
- ("SELECT * FROM testData FULL OUTER JOIN testData2 WHERE key = 2",
classOf[CartesianProduct]),
- ("SELECT * FROM testData JOIN testData2 WHERE key > a",
classOf[CartesianProduct]),
- ("SELECT * FROM testData FULL OUTER JOIN testData2 WHERE key > a",
classOf[CartesianProduct]),
- ("SELECT * FROM testData JOIN testData2 ON key = a",
classOf[SortMergeJoin]),
- ("SELECT * FROM testData JOIN testData2 ON key = a and key = 2",
classOf[SortMergeJoin]),
- ("SELECT * FROM testData JOIN testData2 ON key = a where key = 2",
classOf[SortMergeJoin]),
- ("SELECT * FROM testData LEFT JOIN testData2 ON key = a",
classOf[SortMergeOuterJoin]),
- ("SELECT * FROM testData RIGHT JOIN testData2 ON key = a where key =
2",
- classOf[SortMergeOuterJoin]),
- ("SELECT * FROM testData right join testData2 ON key = a and key =
2",
- classOf[SortMergeOuterJoin]),
- ("SELECT * FROM testData full outer join testData2 ON key = a",
- classOf[SortMergeOuterJoin]),
- ("SELECT * FROM testData left JOIN testData2 ON (key * a != key +
a)",
- classOf[BroadcastNestedLoopJoin]),
- ("SELECT * FROM testData right JOIN testData2 ON (key * a != key +
a)",
- classOf[BroadcastNestedLoopJoin]),
- ("SELECT * FROM testData full JOIN testData2 ON (key * a != key +
a)",
- classOf[BroadcastNestedLoopJoin])
- ).foreach { case (query, joinClass) => assertJoin(query, joinClass) }
+ withSQLConf("spark.sql.autoBroadcastJoinThreshold" -> "0") {
--- End diff --
Some of them will became broadcast joins, should I update the tests without
disabling auto-broadcast?
---
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]
