Merge branch 'SPARK-10117' of github.com:Lewuathe/spark into SPARK-10117

Author: Lewuathe

mllib/src/main/scala/org/apache/spark/ml/source/libsvm/LibSVMRelation.scala

@@ -64,6 +64,10 @@ private[ml] class LibSVMRelation(val path: String, val numFeatures: Int, val vec
 
 }
 
+/**
+ * This is used for creating DataFrame from LibSVM format file.
+ * The LibSVM file path must be specified to DefaultSource.
+ */
 class DefaultSource extends RelationProvider with DataSourceRegister {
 
   override def shortName(): String = "libsvm"

mllib/src/main/scala/org/apache/spark/ml/source/libsvm/package.scala

@@ -68,8 +68,7 @@ public void tearDown() {
 
   @Test
   public void verifyLibSVMDF() {
-    dataset = jsql.read().format("org.apache.spark.ml.source.libsvm").option("vectorType", "dense")
-            .load(path.getPath());
+    dataset = jsql.read().format("libsvm").option("vectorType", "dense").load(path.getPath());
     Assert.assertEquals("label", dataset.columns()[0]);
     Assert.assertEquals("features", dataset.columns()[1]);
     Row r = dataset.first();

mllib/src/test/java/org/apache/spark/ml/source/JavaLibSVMRelationSuite.java

@@ -45,7 +45,7 @@ class LibSVMRelationSuite extends SparkFunSuite with MLlibTestSparkContext {
   }
 
   test("select as sparse vector") {
-    val df = sqlContext.read.libsvm(path)
+    val df = sqlContext.read.format("libsvm").load(path)
     assert(df.columns(0) == "label")
     assert(df.columns(1) == "features")
     val row1 = df.first()
@@ -55,8 +55,8 @@ class LibSVMRelationSuite extends SparkFunSuite with MLlibTestSparkContext {
   }
 
   test("select as dense vector") {
-    val df = sqlContext.read.options(Map("vectorType" -> "dense"))
-      .libsvm(path)
+    val df = sqlContext.read.format("libsvm").options(Map("vectorType" -> "dense"))
+      .load(path)
     assert(df.columns(0) == "label")
     assert(df.columns(1) == "features")
     assert(df.count() == 3)
@@ -75,7 +75,8 @@ class LibSVMRelationSuite extends SparkFunSuite with MLlibTestSparkContext {
     val tempDir = Utils.createTempDir()
     val file = new File(tempDir.getPath, "part-00001")
     Files.write(lines, file, Charsets.US_ASCII)
-    val df = sqlContext.read.option("numFeatures", "100").libsvm(tempDir.toURI.toString)
+    val df = sqlContext.read.option("numFeatures", "100").format("libsvm")
+      .load(tempDir.toURI.toString)
     val row1 = df.first()
     val v = row1.getAs[SparseVector](1)
     assert(v == Vectors.sparse(100, Seq((0, 1.0), (9, 2.0), (19, 3.0), (29, 4.0), (39, 5.0),

mllib/src/test/scala/org/apache/spark/ml/source/LibSVMRelationSuite.scala

@@ -1,3 +1,4 @@
 org.apache.spark.sql.execution.datasources.jdbc.DefaultSource
 org.apache.spark.sql.execution.datasources.json.DefaultSource
 org.apache.spark.sql.execution.datasources.parquet.DefaultSource
+org.apache.spark.ml.source.libsvm.DefaultSource