8 files changed, 108 insertions, 19 deletions
diff --git a/sql/catalyst/pom.xml b/sql/catalyst/pom.xml
index d3f65de16f..fc6d3d7be9 100644
--- a/sql/catalyst/pom.xml
+++ b/sql/catalyst/pom.xml
@@ -65,6 +65,34 @@
         <groupId>org.scalatest</groupId>
         <artifactId>scalatest-maven-plugin</artifactId>
       </plugin>
+
+      <!--
+           This plugin forces the generation of jar containing catalyst test classes,
+           so that the tests classes of external modules can use them. The two execution profiles
+           are necessary - first one for 'mvn package', second one for 'mvn compile'. Ideally,
+           'mvn compile' should not compile test classes and therefore should not need this.
+           However, an open Maven bug (http://jira.codehaus.org/browse/MNG-3559)
+           causes the compilation to fail if catalyst test-jar is not generated. Hence, the
+           second execution profile for 'mvn compile'.
+      -->
+      <plugin>
+        <groupId>org.apache.maven.plugins</groupId>
+        <artifactId>maven-jar-plugin</artifactId>
+        <executions>
+            <execution>
+                <goals>
+                    <goal>test-jar</goal>
+                </goals>
+            </execution>
+            <execution>
+                <id>test-jar-on-compile</id>
+                <phase>compile</phase>
+                <goals>
+                    <goal>test-jar</goal>
+                </goals>
+            </execution>
+        </executions>
+      </plugin>
     </plugins>
   </build>
 </project>
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/HiveTypeCoercion.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/HiveTypeCoercion.scala
index d291814c8a..66bff660ca 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/HiveTypeCoercion.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/HiveTypeCoercion.scala
@@ -22,6 +22,16 @@ import org.apache.spark.sql.catalyst.plans.logical.{LogicalPlan, Project, Union}
 import org.apache.spark.sql.catalyst.rules.Rule
 import org.apache.spark.sql.catalyst.types._
 
+object HiveTypeCoercion {
+  // See https://cwiki.apache.org/confluence/display/Hive/LanguageManual+Types.
+  // The conversion for integral and floating point types have a linear widening hierarchy:
+  val numericPrecedence =
+    Seq(NullType, ByteType, ShortType, IntegerType, LongType, FloatType, DoubleType, DecimalType)
+  // Boolean is only wider than Void
+  val booleanPrecedence = Seq(NullType, BooleanType)
+  val allPromotions: Seq[Seq[DataType]] = numericPrecedence :: booleanPrecedence :: Nil
+}
+
 /**
  * A collection of [[catalyst.rules.Rule Rules]] that can be used to coerce differing types that
  * participate in operations into compatible ones.  Most of these rules are based on Hive semantics,
@@ -116,19 +126,18 @@ trait HiveTypeCoercion {
    *
    * Additionally, all types when UNION-ed with strings will be promoted to strings.
    * Other string conversions are handled by PromoteStrings.
+   *
+   * Widening types might result in loss of precision in the following cases:
+   * - IntegerType to FloatType
+   * - LongType to FloatType
+   * - LongType to DoubleType
    */
   object WidenTypes extends Rule[LogicalPlan] {
-    // See https://cwiki.apache.org/confluence/display/Hive/LanguageManual+Types.
-    // The conversion for integral and floating point types have a linear widening hierarchy:
-    val numericPrecedence =
-      Seq(NullType, ByteType, ShortType, IntegerType, LongType, FloatType, DoubleType, DecimalType)
-    // Boolean is only wider than Void
-    val booleanPrecedence = Seq(NullType, BooleanType)
-    val allPromotions: Seq[Seq[DataType]] = numericPrecedence :: booleanPrecedence :: Nil
 
     def findTightestCommonType(t1: DataType, t2: DataType): Option[DataType] = {
       // Try and find a promotion rule that contains both types in question.
-      val applicableConversion = allPromotions.find(p => p.contains(t1) && p.contains(t2))
+      val applicableConversion =
+        HiveTypeCoercion.allPromotions.find(p => p.contains(t1) && p.contains(t2))
 
       // If found return the widest common type, otherwise None
       applicableConversion.map(_.filter(t => t == t1 || t == t2).last)
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/QueryPlan.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/QueryPlan.scala
index 8199a80f5d..00e2d3bc24 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/QueryPlan.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/QueryPlan.scala
@@ -18,7 +18,9 @@
 package org.apache.spark.sql.catalyst.plans
 
 import org.apache.spark.sql.catalyst.expressions.{Attribute, Expression}
+import org.apache.spark.sql.catalyst.plans
 import org.apache.spark.sql.catalyst.trees.TreeNode
+import org.apache.spark.sql.catalyst.types.{ArrayType, DataType, StructField, StructType}
 
 abstract class QueryPlan[PlanType <: TreeNode[PlanType]] extends TreeNode[PlanType] {
   self: PlanType with Product =>
@@ -123,4 +125,53 @@ abstract class QueryPlan[PlanType <: TreeNode[PlanType]] extends TreeNode[PlanTy
       case other => Nil
     }.toSeq
   }
+
+  protected def generateSchemaString(schema: Seq[Attribute]): String = {
+    val builder = new StringBuilder
+    builder.append("root\n")
+    val prefix = " |"
+    schema.foreach { attribute =>
+      val name = attribute.name
+      val dataType = attribute.dataType
+      dataType match {
+        case fields: StructType =>
+          builder.append(s"$prefix-- $name: $StructType\n")
+          generateSchemaString(fields, s"$prefix    |", builder)
+        case ArrayType(fields: StructType) =>
+          builder.append(s"$prefix-- $name: $ArrayType[$StructType]\n")
+          generateSchemaString(fields, s"$prefix    |", builder)
+        case ArrayType(elementType: DataType) =>
+          builder.append(s"$prefix-- $name: $ArrayType[$elementType]\n")
+        case _ => builder.append(s"$prefix-- $name: $dataType\n")
+      }
+    }
+
+    builder.toString()
+  }
+
+  protected def generateSchemaString(
+      schema: StructType,
+      prefix: String,
+      builder: StringBuilder): StringBuilder = {
+    schema.fields.foreach {
+      case StructField(name, fields: StructType, _) =>
+        builder.append(s"$prefix-- $name: $StructType\n")
+        generateSchemaString(fields, s"$prefix    |", builder)
+      case StructField(name, ArrayType(fields: StructType), _) =>
+        builder.append(s"$prefix-- $name: $ArrayType[$StructType]\n")
+        generateSchemaString(fields, s"$prefix    |", builder)
+      case StructField(name, ArrayType(elementType: DataType), _) =>
+        builder.append(s"$prefix-- $name: $ArrayType[$elementType]\n")
+      case StructField(name, fieldType: DataType, _) =>
+        builder.append(s"$prefix-- $name: $fieldType\n")
+    }
+
+    builder
+  }
+
+  /** Returns the output schema in the tree format. */
+  def schemaString: String = generateSchemaString(output)
+
+  /** Prints out the schema in the tree format */
+  def printSchema(): Unit = println(schemaString)
 }
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/CombiningLimitsSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/CombiningLimitsSuite.scala
index 714f01843c..4896f1b955 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/CombiningLimitsSuite.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/CombiningLimitsSuite.scala
@@ -18,11 +18,12 @@
 package org.apache.spark.sql.catalyst.optimizer
 
 import org.apache.spark.sql.catalyst.plans.logical._
+import org.apache.spark.sql.catalyst.plans.PlanTest
 import org.apache.spark.sql.catalyst.rules._
 import org.apache.spark.sql.catalyst.dsl.plans._
 import org.apache.spark.sql.catalyst.dsl.expressions._
 
-class CombiningLimitsSuite extends OptimizerTest {
+class CombiningLimitsSuite extends PlanTest {
 
   object Optimize extends RuleExecutor[LogicalPlan] {
     val batches =
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/ConstantFoldingSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/ConstantFoldingSuite.scala
index 6efc0e211e..cea97c584f 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/ConstantFoldingSuite.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/ConstantFoldingSuite.scala
@@ -20,6 +20,7 @@ package org.apache.spark.sql.catalyst.optimizer
 import org.apache.spark.sql.catalyst.analysis.EliminateAnalysisOperators
 import org.apache.spark.sql.catalyst.expressions._
 import org.apache.spark.sql.catalyst.plans.logical.{LocalRelation, LogicalPlan}
+import org.apache.spark.sql.catalyst.plans.PlanTest
 import org.apache.spark.sql.catalyst.rules.RuleExecutor
 import org.apache.spark.sql.catalyst.types._
 
@@ -27,7 +28,7 @@ import org.apache.spark.sql.catalyst.types._
 import org.apache.spark.sql.catalyst.dsl.plans._
 import org.apache.spark.sql.catalyst.dsl.expressions._
 
-class ConstantFoldingSuite extends OptimizerTest {
+class ConstantFoldingSuite extends PlanTest {
 
   object Optimize extends RuleExecutor[LogicalPlan] {
     val batches =
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/FilterPushdownSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/FilterPushdownSuite.scala
index 1f67c80e54..ebb123c1f9 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/FilterPushdownSuite.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/FilterPushdownSuite.scala
@@ -20,13 +20,12 @@ package org.apache.spark.sql.catalyst.optimizer
 import org.apache.spark.sql.catalyst.analysis
 import org.apache.spark.sql.catalyst.analysis.EliminateAnalysisOperators
 import org.apache.spark.sql.catalyst.plans.logical._
-import org.apache.spark.sql.catalyst.plans.LeftOuter
-import org.apache.spark.sql.catalyst.plans.RightOuter
+import org.apache.spark.sql.catalyst.plans.{PlanTest, LeftOuter, RightOuter}
 import org.apache.spark.sql.catalyst.rules._
 import org.apache.spark.sql.catalyst.dsl.plans._
 import org.apache.spark.sql.catalyst.dsl.expressions._
 
-class FilterPushdownSuite extends OptimizerTest {
+class FilterPushdownSuite extends PlanTest {
 
   object Optimize extends RuleExecutor[LogicalPlan] {
     val batches =
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/SimplifyCaseConversionExpressionsSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/SimplifyCaseConversionExpressionsSuite.scala
index df1409fe7b..22992fb6f5 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/SimplifyCaseConversionExpressionsSuite.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/SimplifyCaseConversionExpressionsSuite.scala
@@ -19,13 +19,14 @@ package org.apache.spark.sql.catalyst.optimizer
 
 import org.apache.spark.sql.catalyst.expressions._
 import org.apache.spark.sql.catalyst.plans.logical._
+import org.apache.spark.sql.catalyst.plans.PlanTest
 import org.apache.spark.sql.catalyst.rules._
 
 /* Implicit conversions */
 import org.apache.spark.sql.catalyst.dsl.expressions._
 import org.apache.spark.sql.catalyst.dsl.plans._
 
-class SimplifyCaseConversionExpressionsSuite extends OptimizerTest {
+class SimplifyCaseConversionExpressionsSuite extends PlanTest {
 
   object Optimize extends RuleExecutor[LogicalPlan] {
     val batches =
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/OptimizerTest.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/plans/PlanTest.scala
index 89982d5cd8..7e9f47ef21 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/OptimizerTest.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/plans/PlanTest.scala
@@ -15,19 +15,18 @@
  * limitations under the License.
  */
 
-package org.apache.spark.sql.catalyst.optimizer
+package org.apache.spark.sql.catalyst.plans
 
 import org.scalatest.FunSuite
 
-import org.apache.spark.sql.catalyst.expressions._
+import org.apache.spark.sql.catalyst.expressions.{ExprId, AttributeReference}
 import org.apache.spark.sql.catalyst.plans.logical.LogicalPlan
 import org.apache.spark.sql.catalyst.util._
 
 /**
- * Provides helper methods for comparing plans produced by optimization rules with the expected
- * result
+ * Provides helper methods for comparing plans.
  */
-class OptimizerTest extends FunSuite {
+class PlanTest extends FunSuite {
 
   /**
    * Since attribute references are given globally unique ids during analysis,