From 511a7314037219c23e824ea5363bf7f1df55bab3 Mon Sep 17 00:00:00 2001
From: Michael Armbrust <michael@databricks.com>
Date: Mon, 21 Jul 2014 18:18:17 -0700
Subject: [PATCH] [SPARK-2561][SQL] Fix apply schema
We need to use the analyzed attributes otherwise we end up with a tree that will never resolve.
Author: Michael Armbrust <michael@databricks.com>
Closes #1470 from marmbrus/fixApplySchema and squashes the following commits:
f968195 [Michael Armbrust] Use analyzed attributes when applying the schema.
4969015 [Michael Armbrust] Add test case.
---
.../src/main/scala/org/apache/spark/sql/SchemaRDD.scala | 2 +-
.../src/test/scala/org/apache/spark/sql/DslQuerySuite.scala | 6 ++++++
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/sql/core/src/main/scala/org/apache/spark/sql/SchemaRDD.scala b/sql/core/src/main/scala/org/apache/spark/sql/SchemaRDD.scala
index 993d085c75..31d27bb4f0 100644
--- a/sql/core/src/main/scala/org/apache/spark/sql/SchemaRDD.scala
+++ b/sql/core/src/main/scala/org/apache/spark/sql/SchemaRDD.scala
@@ -430,7 +430,7 @@ class SchemaRDD(
* @group schema
*/
private def applySchema(rdd: RDD[Row]): SchemaRDD = {
- new SchemaRDD(sqlContext, SparkLogicalPlan(ExistingRdd(logicalPlan.output, rdd)))
+ new SchemaRDD(sqlContext, SparkLogicalPlan(ExistingRdd(queryExecution.analyzed.output, rdd)))
}
// =======================================================================
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DslQuerySuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DslQuerySuite.scala
index 68dae58728..c8ea01c4e1 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/DslQuerySuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/DslQuerySuite.scala
@@ -33,6 +33,12 @@ class DslQuerySuite extends QueryTest {
testData.collect().toSeq)
}
+ test("repartition") {
+ checkAnswer(
+ testData.select('key).repartition(10).select('key),
+ testData.select('key).collect().toSeq)
+ }
+
test("agg") {
checkAnswer(
testData2.groupBy('a)('a, Sum('b)),
--
GitLab