Spark 2.0.0 - SQL - Running query with outer join from 1.6 fails

[Don Drake <do...@gmail.com>]

So I was able to reproduce in a simple case the issue I'm seeing with a
query from Spark 1.6.2 that would run fine that is no longer working on
Spark 2.0.

Example code:
https://gist.github.com/dondrake/c136d61503b819f0643f8c02854a9cdf

Here's the code for Spark 2.0 that doesn't run (this runs fine in Spark
1.6.2):

case class C1(f1: String, f2: String, f3: String, f4: String)
case class C2(g1: String, g2: String, g3: String, g4: String)
case class C3(h1: String, h2: String, h3: String, h4: String)

val sqlContext = spark.sqlContext

val c1 = sc.parallelize(Seq(
  C1("h1", "c1a1", "c1b1", "c1c1"),
  C1("h2", "c1a2", "c1b2", "c1c2"),
  C1(null, "c1a3", "c1b3", "c1c3")
  )).toDF
c1.createOrReplaceTempView("c1")

val c2 = sc.parallelize(Seq(
  C2("h1", "c2a1", "c2b1", "c2c1"),
  C2("h2", "c2a2", "c2b2", "c2c2"),
  C2(null, "c2a3", "c2b3", "c2c3"),
  C2(null, "c2a4", "c2b4", "c2c4"),
  C2("h333", "c2a333", "c2b333", "c2c333")
  )).toDF
c2.createOrReplaceTempView("c2")

val c3 = sc.parallelize(Seq(
  C3("h1", "c3a1", "c3b1", "c3c1"),
  C3("h2", "c3a2", "c3b2", "c3c2"),
  C3(null, "c3a3", "c3b3", "c3c3")
  )).toDF
c3.createOrReplaceTempView("c3")

// doesn't work in Spark 2.0, works in Spark 1.6
val bad_df = sqlContext.sql("""
  select *
  from c1, c3
  left outer join c2 on (c1.f1 = c2.g1)
  where c1.f1 = c3.h1
""").show()

// works in both
val works_df = sqlContext.sql("""
  select *
  from c1
  left outer join c2 on (c1.f1 = c2.g1),
  c3
  where c1.f1 = c3.h1
""").show()


Here's the output after running bad_df in Spark 2.0:

scala> val bad_df = sqlContext.sql("""
     |   select *
     |   from c1, c3
     |   left outer join c2 on (c1.f1 = c2.g1)
     |   where c1.f1 = c3.h1
     | """).show()
org.apache.spark.sql.AnalysisException: cannot resolve '`c1.f1`' given
input columns: [h3, g3, h4, g2, g4, h2, h1, g1]; line 4 pos 25
  at
org.apache.spark.sql.catalyst.analysis.package$AnalysisErrorAt.failAnalysis(package.scala:42)
  at
org.apache.spark.sql.catalyst.analysis.CheckAnalysis$$anonfun$checkAnalysis$1$$anonfun$apply$2.applyOrElse(CheckAnalysis.scala:77)
  at
org.apache.spark.sql.catalyst.analysis.CheckAnalysis$$anonfun$checkAnalysis$1$$anonfun$apply$2.applyOrElse(CheckAnalysis.scala:74)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$transformUp$1.apply(TreeNode.scala:301)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$transformUp$1.apply(TreeNode.scala:301)
  at
org.apache.spark.sql.catalyst.trees.CurrentOrigin$.withOrigin(TreeNode.scala:69)
  at
org.apache.spark.sql.catalyst.trees.TreeNode.transformUp(TreeNode.scala:300)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$4.apply(TreeNode.scala:298)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$4.apply(TreeNode.scala:298)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$5.apply(TreeNode.scala:321)
  at
org.apache.spark.sql.catalyst.trees.TreeNode.mapProductIterator(TreeNode.scala:179)
  at
org.apache.spark.sql.catalyst.trees.TreeNode.transformChildren(TreeNode.scala:319)
  at
org.apache.spark.sql.catalyst.trees.TreeNode.transformUp(TreeNode.scala:298)
  at
org.apache.spark.sql.catalyst.plans.QueryPlan.transformExpressionUp$1(QueryPlan.scala:190)
  at org.apache.spark.sql.catalyst.plans.QueryPlan.org
$apache$spark$sql$catalyst$plans$QueryPlan$$recursiveTransform$2(QueryPlan.scala:201)
  at
org.apache.spark.sql.catalyst.plans.QueryPlan$$anonfun$5.apply(QueryPlan.scala:209)
  at
org.apache.spark.sql.catalyst.trees.TreeNode.mapProductIterator(TreeNode.scala:179)
  at
org.apache.spark.sql.catalyst.plans.QueryPlan.transformExpressionsUp(QueryPlan.scala:209)
  at
org.apache.spark.sql.catalyst.analysis.CheckAnalysis$$anonfun$checkAnalysis$1.apply(CheckAnalysis.scala:74)
  at
org.apache.spark.sql.catalyst.analysis.CheckAnalysis$$anonfun$checkAnalysis$1.apply(CheckAnalysis.scala:67)
  at
org.apache.spark.sql.catalyst.trees.TreeNode.foreachUp(TreeNode.scala:126)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$foreachUp$1.apply(TreeNode.scala:125)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$foreachUp$1.apply(TreeNode.scala:125)
  at scala.collection.immutable.List.foreach(List.scala:381)
  at
org.apache.spark.sql.catalyst.trees.TreeNode.foreachUp(TreeNode.scala:125)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$foreachUp$1.apply(TreeNode.scala:125)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$foreachUp$1.apply(TreeNode.scala:125)
  at scala.collection.immutable.List.foreach(List.scala:381)
  at
org.apache.spark.sql.catalyst.trees.TreeNode.foreachUp(TreeNode.scala:125)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$foreachUp$1.apply(TreeNode.scala:125)
  at
org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$foreachUp$1.apply(TreeNode.scala:125)
  at scala.collection.immutable.List.foreach(List.scala:381)
  at
org.apache.spark.sql.catalyst.trees.TreeNode.foreachUp(TreeNode.scala:125)
  at
org.apache.spark.sql.catalyst.analysis.CheckAnalysis$class.checkAnalysis(CheckAnalysis.scala:67)
  at
org.apache.spark.sql.catalyst.analysis.Analyzer.checkAnalysis(Analyzer.scala:58)
  at
org.apache.spark.sql.execution.QueryExecution.assertAnalyzed(QueryExecution.scala:49)
  at org.apache.spark.sql.Dataset$.ofRows(Dataset.scala:64)
  at org.apache.spark.sql.SparkSession.sql(SparkSession.scala:582)
  at org.apache.spark.sql.SQLContext.sql(SQLContext.scala:682)
  ... 53 elided

scala>

I confirmed this fails on the Spark 2.0 nightly build as well.

Thoughts? Should I open a JIRA?

-Don
-- 
Donald Drake
Drake Consulting
http://www.drakeconsulting.com/
https://twitter.com/dondrake <http://www.MailLaunder.com/>
800-733-2143