fixed a couple of format / error bugs

Author: JDrit

sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/ddl.scala

@@ -22,6 +22,7 @@ import java.util.ServiceLoader
 import scala.collection.Iterator
 import scala.collection.JavaConversions._
 import scala.language.{existentials, implicitConversions}
+import scala.util.{Failure, Success, Try}
 import scala.util.matching.Regex
 
 import org.apache.hadoop.fs.Path
@@ -196,26 +197,27 @@ private[sql] class DDLParser(
 
 private[sql] object ResolvedDataSource extends Logging {
 
-  /** Tries to load the particular class */
-  private def tryLoad(loader: ClassLoader, provider: String): Option[Class[_]] = try {
-    Some(loader.loadClass(provider))
-  } catch {
-    case cnf: ClassNotFoundException => None
-  }
-
   /** Given a provider name, look up the data source class definition. */
   def lookupDataSource(provider: String): Class[_] = {
+    val provider2 = s"$provider.DefaultSource"
     val loader = Utils.getContextOrSparkClassLoader
     val serviceLoader = ServiceLoader.load(classOf[DataSourceRegister], loader)
 
-    serviceLoader.iterator().filter(_.format() == provider).toList match {
-      case Nil => tryLoad(loader, provider).orElse(tryLoad(loader, s"$provider.DefaultSource"))
-        .getOrElse(if (provider.startsWith("org.apache.spark.sql.hive.orc")) {
-            sys.error("The ORC data source must be used with Hive support enabled.")
-          } else {
-            sys.error(s"Failed to load class for data source: $provider")
-          })
+    serviceLoader.iterator().filter(_.format().equalsIgnoreCase(provider)).toList match {
+      /** the provider format did not match any given registered aliases */
+      case Nil => Try(loader.loadClass(provider)).orElse(Try(loader.loadClass(provider2))) match {
+        case Success(dataSource) => dataSource
+        case Failure(error) => if (provider.startsWith("org.apache.spark.sql.hive.orc")) {
+          throw new ClassNotFoundException(
+            "The ORC data source must be used with Hive support enabled.", error)
+        } else {
+          throw new ClassNotFoundException(
+            s"Failed to load class for data source: $provider", error)
+        }
+      }
+      /** there is exactly one registered alias */
       case head :: Nil => head.getClass
+      /** There are multiple registered aliases for the input */
       case sources => sys.error(s"Multiple sources found for $provider, " +
         s"(${sources.map(_.getClass.getName).mkString(", ")}), " +
         "please specify the fully qualified class name")

sql/core/src/test/scala/org/apache/spark/sql/sources/DDLSourceLoadSuite.scala

@@ -76,4 +76,10 @@ class DDLSourceLoadSuite extends DataSourceTest {
     caseInsensitiveContext.read.format("org.apache.spark.sql.sources.FakeSourceOne")
       .load().schema == StructType(Seq(StructField("stringType", StringType, nullable = false)))
   }
+
+  test("Loading Orc") {
+    intercept[ClassNotFoundException] {
+      caseInsensitiveContext.read.format("orc").load()
+    }
+  }
 }

sql/hive/src/main/resources/META-INF/services/org.apache.spark.sql.sources.DataSourceRegister

@@ -1 +1 @@
-org.apache.spark.sql.hive.orc.DefaultSource
+org.apache.spark.sql.hive.orc.DefaultSource