[SPARK-9580] [SQL] Replace singletons in SQL tests

A fundamental limitation of the existing SQL tests is that *there is simply no way to create your own `SparkContext`*. This is a serious limitation because the user may wish to use a different master or config. As a case in point, `BroadcastJoinSuite` is entirely commented out because there is no way to make it pass with the existing infrastructure.

This patch removes the singletons `TestSQLContext` and `TestData`, and instead introduces a `SharedSQLContext` that starts a context per suite. Unfortunately the singletons were so ingrained in the SQL tests that this patch necessarily needed to touch *all* the SQL test files.

<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/apache/spark/8111)
<!-- Reviewable:end -->

Author: Andrew Or <andrew@databricks.com>

Closes #8111 from andrewor14/sql-tests-refactor.

14 files changed, 36 insertions, 33 deletions

diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveMetastoreCatalogSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveMetastoreCatalogSuite.scala
index 59e65ff97b..574624d501 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveMetastoreCatalogSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveMetastoreCatalogSuite.scala
@@ -26,7 +26,7 @@ import org.apache.spark.sql.hive.test.TestHive.implicits._
 import org.apache.spark.sql.sources.DataSourceTest
 import org.apache.spark.sql.test.{ExamplePointUDT, SQLTestUtils}
 import org.apache.spark.sql.types.{DecimalType, StringType, StructType}
-import org.apache.spark.sql.{Row, SaveMode}
+import org.apache.spark.sql.{Row, SaveMode, SQLContext}
 import org.apache.spark.{Logging, SparkFunSuite}
 
 
@@ -53,7 +53,8 @@ class HiveMetastoreCatalogSuite extends SparkFunSuite with Logging {
 }
 
 class DataSourceWithHiveMetastoreCatalogSuite extends DataSourceTest with SQLTestUtils {
-  override val sqlContext = TestHive
+  override def _sqlContext: SQLContext = TestHive
+  import testImplicits._
 
   private val testDF = range(1, 3).select(
     ('id + 0.1) cast DecimalType(10, 3) as 'd1,
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveParquetSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveParquetSuite.scala
index 1fa005d5f9..fe0db5228d 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveParquetSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveParquetSuite.scala
@@ -19,14 +19,13 @@ package org.apache.spark.sql.hive
 
 import org.apache.spark.sql.hive.test.TestHive
 import org.apache.spark.sql.execution.datasources.parquet.ParquetTest
-import org.apache.spark.sql.{QueryTest, Row}
+import org.apache.spark.sql.{QueryTest, Row, SQLContext}
 
 case class Cases(lower: String, UPPER: String)
 
 class HiveParquetSuite extends QueryTest with ParquetTest {
-  val sqlContext = TestHive
-
-  import sqlContext._
+  private val ctx = TestHive
+  override def _sqlContext: SQLContext = ctx
 
   test("Case insensitive attribute names") {
     withParquetTable((1 to 4).map(i => Cases(i.toString, i.toString)), "cases") {
@@ -54,7 +53,7 @@ class HiveParquetSuite extends QueryTest with ParquetTest {
   test("Converting Hive to Parquet Table via saveAsParquetFile") {
     withTempPath { dir =>
       sql("SELECT * FROM src").write.parquet(dir.getCanonicalPath)
-      read.parquet(dir.getCanonicalPath).registerTempTable("p")
+      ctx.read.parquet(dir.getCanonicalPath).registerTempTable("p")
       withTempTable("p") {
         checkAnswer(
           sql("SELECT * FROM src ORDER BY key"),
@@ -67,7 +66,7 @@ class HiveParquetSuite extends QueryTest with ParquetTest {
     withParquetTable((1 to 10).map(i => (i, s"val_$i")), "t") {
       withTempPath { file =>
         sql("SELECT * FROM t LIMIT 1").write.parquet(file.getCanonicalPath)
-        read.parquet(file.getCanonicalPath).registerTempTable("p")
+        ctx.read.parquet(file.getCanonicalPath).registerTempTable("p")
         withTempTable("p") {
           // let's do three overwrites for good measure
           sql("INSERT OVERWRITE TABLE p SELECT * FROM t")
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/MetastoreDataSourcesSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/MetastoreDataSourcesSuite.scala
index 7f36a483a3..20a50586d5 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/MetastoreDataSourcesSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/MetastoreDataSourcesSuite.scala
@@ -22,7 +22,6 @@ import java.io.{IOException, File}
 import scala.collection.mutable.ArrayBuffer
 
 import org.apache.hadoop.fs.Path
-import org.apache.hadoop.mapred.InvalidInputException
 import org.scalatest.BeforeAndAfterAll
 
 import org.apache.spark.Logging
@@ -42,7 +41,8 @@ import org.apache.spark.util.Utils
  */
 class MetastoreDataSourcesSuite extends QueryTest with SQLTestUtils with BeforeAndAfterAll
   with Logging {
-  override val sqlContext = TestHive
+  override def _sqlContext: SQLContext = TestHive
+  private val sqlContext = _sqlContext
 
   var jsonFilePath: String = _
 
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/MultiDatabaseSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/MultiDatabaseSuite.scala
index 73852f13ad..417e8b0791 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/MultiDatabaseSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/MultiDatabaseSuite.scala
@@ -22,9 +22,8 @@ import org.apache.spark.sql.test.SQLTestUtils
 import org.apache.spark.sql.{QueryTest, SQLContext, SaveMode}
 
 class MultiDatabaseSuite extends QueryTest with SQLTestUtils {
-  override val sqlContext: SQLContext = TestHive
-
-  import sqlContext.sql
+  override val _sqlContext: SQLContext = TestHive
+  private val sqlContext = _sqlContext
 
   private val df = sqlContext.range(10).coalesce(1)
 
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/ParquetHiveCompatibilitySuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/ParquetHiveCompatibilitySuite.scala
index 251e0324bf..13452e71a1 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/ParquetHiveCompatibilitySuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/ParquetHiveCompatibilitySuite.scala
@@ -26,7 +26,8 @@ import org.apache.spark.sql.{Row, SQLConf, SQLContext}
 class ParquetHiveCompatibilitySuite extends ParquetCompatibilityTest {
   import ParquetCompatibilityTest.makeNullable
 
-  override val sqlContext: SQLContext = TestHive
+  override def _sqlContext: SQLContext = TestHive
+  private val sqlContext = _sqlContext
 
   /**
    * Set the staging directory (and hence path to ignore Parquet files under)
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/UDFSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/UDFSuite.scala
index 9b3ede43ee..7ee1c8d13a 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/UDFSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/UDFSuite.scala
@@ -17,14 +17,12 @@
 
 package org.apache.spark.sql.hive
 
-import org.apache.spark.sql.{Row, QueryTest}
+import org.apache.spark.sql.QueryTest
 
 case class FunctionResult(f1: String, f2: String)
 
 class UDFSuite extends QueryTest {
-
   private lazy val ctx = org.apache.spark.sql.hive.test.TestHive
-  import ctx.implicits._
 
   test("UDF case insensitive") {
     ctx.udf.register("random0", () => { Math.random() })
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/AggregationQuerySuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/AggregationQuerySuite.scala
index 7b5aa4763f..a312f84958 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/AggregationQuerySuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/AggregationQuerySuite.scala
@@ -17,17 +17,18 @@
 
 package org.apache.spark.sql.hive.execution
 
+import org.scalatest.BeforeAndAfterAll
+
+import org.apache.spark.sql._
 import org.apache.spark.sql.execution.aggregate
 import org.apache.spark.sql.hive.test.TestHive
 import org.apache.spark.sql.test.SQLTestUtils
 import org.apache.spark.sql.types.{IntegerType, StringType, StructField, StructType}
-import org.apache.spark.sql._
-import org.scalatest.BeforeAndAfterAll
 import _root_.test.org.apache.spark.sql.hive.aggregate.{MyDoubleAvg, MyDoubleSum}
 
 abstract class AggregationQuerySuite extends QueryTest with SQLTestUtils with BeforeAndAfterAll {
-
-  override val sqlContext = TestHive
+  override def _sqlContext: SQLContext = TestHive
+  protected val sqlContext = _sqlContext
   import sqlContext.implicits._
 
   var originalUseAggregate2: Boolean = _
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveExplainSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveExplainSuite.scala
index 44c5b80392..11d7a872df 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveExplainSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveExplainSuite.scala
@@ -26,8 +26,8 @@ import org.apache.spark.sql.test.SQLTestUtils
  * A set of tests that validates support for Hive Explain command.
  */
 class HiveExplainSuite extends QueryTest with SQLTestUtils {
-
-  def sqlContext: SQLContext = TestHive
+  override def _sqlContext: SQLContext = TestHive
+  private val sqlContext = _sqlContext
 
   test("explain extended command") {
     checkExistence(sql(" explain   select * from src where key=123 "), true,
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/SQLQuerySuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/SQLQuerySuite.scala
index 79a136ae6f..8b8f520776 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/SQLQuerySuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/SQLQuerySuite.scala
@@ -66,7 +66,8 @@ class MyDialect extends DefaultParserDialect
  * valid, but Hive currently cannot execute it.
  */
 class SQLQuerySuite extends QueryTest with SQLTestUtils {
-  override def sqlContext: SQLContext = TestHive
+  override def _sqlContext: SQLContext = TestHive
+  private val sqlContext = _sqlContext
 
   test("UDTF") {
     sql(s"ADD JAR ${TestHive.getHiveFile("TestUDTF.jar").getCanonicalPath()}")
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/ScriptTransformationSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/ScriptTransformationSuite.scala
index 0875232aed..9aca40f15a 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/ScriptTransformationSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/ScriptTransformationSuite.scala
@@ -31,7 +31,8 @@ import org.apache.spark.sql.types.StringType
 
 class ScriptTransformationSuite extends SparkPlanTest {
 
-  override def sqlContext: SQLContext = TestHive
+  override def _sqlContext: SQLContext = TestHive
+  private val sqlContext = _sqlContext
 
   private val noSerdeIOSchema = HiveScriptIOSchema(
     inputRowFormat = Seq.empty,
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/orc/OrcTest.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/orc/OrcTest.scala
index 145965388d..f7ba20ff41 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/orc/OrcTest.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/orc/OrcTest.scala
@@ -27,8 +27,8 @@ import org.apache.spark.sql._
 import org.apache.spark.sql.test.SQLTestUtils
 
 private[sql] trait OrcTest extends SQLTestUtils { this: SparkFunSuite =>
-  lazy val sqlContext = org.apache.spark.sql.hive.test.TestHive
-
+  protected override def _sqlContext: SQLContext = org.apache.spark.sql.hive.test.TestHive
+  protected val sqlContext = _sqlContext
   import sqlContext.implicits._
   import sqlContext.sparkContext
 
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/parquetSuites.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/parquetSuites.scala
index 50f02432da..34d3434569 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/parquetSuites.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/parquetSuites.scala
@@ -685,7 +685,8 @@ class ParquetSourceSuite extends ParquetPartitioningTest {
  * A collection of tests for parquet data with various forms of partitioning.
  */
 abstract class ParquetPartitioningTest extends QueryTest with SQLTestUtils with BeforeAndAfterAll {
-  override def sqlContext: SQLContext = TestHive
+  override def _sqlContext: SQLContext = TestHive
+  protected val sqlContext = _sqlContext
 
   var partitionedTableDir: File = null
   var normalTableDir: File = null
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/sources/CommitFailureTestRelationSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/sources/CommitFailureTestRelationSuite.scala
index e976125b37..b4640b1616 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/sources/CommitFailureTestRelationSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/sources/CommitFailureTestRelationSuite.scala
@@ -18,14 +18,16 @@
 package org.apache.spark.sql.sources
 
 import org.apache.hadoop.fs.Path
-import org.apache.spark.deploy.SparkHadoopUtil
 import org.apache.spark.{SparkException, SparkFunSuite}
+import org.apache.spark.deploy.SparkHadoopUtil
+import org.apache.spark.sql.SQLContext
 import org.apache.spark.sql.hive.test.TestHive
 import org.apache.spark.sql.test.SQLTestUtils
 
 
 class CommitFailureTestRelationSuite extends SparkFunSuite with SQLTestUtils {
-  override val sqlContext = TestHive
+  override def _sqlContext: SQLContext = TestHive
+  private val sqlContext = _sqlContext
 
   // When committing a task, `CommitFailureTestSource` throws an exception for testing purpose.
   val dataSourceName: String = classOf[CommitFailureTestSource].getCanonicalName
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/sources/hadoopFsRelationSuites.scala b/sql/hive/src/test/scala/org/apache/spark/sql/sources/hadoopFsRelationSuites.scala
index 2a69d331b6..af445626fb 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/sources/hadoopFsRelationSuites.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/sources/hadoopFsRelationSuites.scala
@@ -34,9 +34,8 @@ import org.apache.spark.sql.types._
 
 
 abstract class HadoopFsRelationTest extends QueryTest with SQLTestUtils {
-  override lazy val sqlContext: SQLContext = TestHive
-
-  import sqlContext.sql
+  override def _sqlContext: SQLContext = TestHive
+  protected val sqlContext = _sqlContext
   import sqlContext.implicits._
 
   val dataSourceName: String