From 023ed194b465b1ada80be1ef0deea0da99ce109e Mon Sep 17 00:00:00 2001 From: Matei Zaharia <matei@eecs.berkeley.edu> Date: Sat, 16 Oct 2010 21:21:16 -0700 Subject: [PATCH] Fixed some whitespace --- src/scala/spark/HadoopFile.scala | 4 ++-- src/scala/spark/HttpServer.scala | 2 +- src/scala/spark/RDD.scala | 22 +++++++++++----------- 3 files changed, 14 insertions(+), 14 deletions(-) diff --git a/src/scala/spark/HadoopFile.scala b/src/scala/spark/HadoopFile.scala index aa28569eda..5746c433ee 100644 --- a/src/scala/spark/HadoopFile.scala +++ b/src/scala/spark/HadoopFile.scala @@ -15,7 +15,7 @@ import org.apache.hadoop.util.ReflectionUtils /** A Spark split class that wraps around a Hadoop InputSplit */ @serializable class HadoopSplit(@transient s: InputSplit) -extends Split { +extends Split { val inputSplit = new SerializableWritable[InputSplit](s) // Hadoop gives each split a unique toString value, so use this as our ID @@ -48,7 +48,7 @@ extends RDD[(K, V)](sc) { } override def splits = splits_ - + override def iterator(theSplit: Split) = new Iterator[(K, V)] { val split = theSplit.asInstanceOf[HadoopSplit] var reader: RecordReader[K, V] = null diff --git a/src/scala/spark/HttpServer.scala b/src/scala/spark/HttpServer.scala index 08eb08bbe3..d5bdd245bb 100644 --- a/src/scala/spark/HttpServer.scala +++ b/src/scala/spark/HttpServer.scala @@ -11,7 +11,7 @@ import org.eclipse.jetty.util.thread.QueuedThreadPool /** - * Exception type thrown by HttpServer when it is in the wrong state + * Exception type thrown by HttpServer when it is in the wrong state * for an operation. */ class ServerStateException(message: String) extends Exception(message) diff --git a/src/scala/spark/RDD.scala b/src/scala/spark/RDD.scala index 20865d7d28..9dd8bc9dce 100644 --- a/src/scala/spark/RDD.scala +++ b/src/scala/spark/RDD.scala @@ -78,7 +78,7 @@ abstract class RDD[T: ClassManifest]( case _ => throw new UnsupportedOperationException("empty collection") } - def count(): Long = + def count(): Long = try { map(x => 1L).reduce(_+_) } catch { case e: UnsupportedOperationException => 0L } @@ -128,7 +128,7 @@ extends RDDTask[Option[T], T](rdd, split) with Logging { } class MappedRDD[U: ClassManifest, T: ClassManifest]( - prev: RDD[T], f: T => U) + prev: RDD[T], f: T => U) extends RDD[U](prev.sparkContext) { override def splits = prev.splits override def preferredLocations(split: Split) = prev.preferredLocations(split) @@ -137,7 +137,7 @@ extends RDD[U](prev.sparkContext) { } class FilteredRDD[T: ClassManifest]( - prev: RDD[T], f: T => Boolean) + prev: RDD[T], f: T => Boolean) extends RDD[T](prev.sparkContext) { override def splits = prev.splits override def preferredLocations(split: Split) = prev.preferredLocations(split) @@ -146,7 +146,7 @@ extends RDD[T](prev.sparkContext) { } class FlatMappedRDD[U: ClassManifest, T: ClassManifest]( - prev: RDD[T], f: T => Traversable[U]) + prev: RDD[T], f: T => Traversable[U]) extends RDD[U](prev.sparkContext) { override def splits = prev.splits override def preferredLocations(split: Split) = prev.preferredLocations(split) @@ -155,7 +155,7 @@ extends RDD[U](prev.sparkContext) { override def taskStarted(split: Split, slot: SlaveOffer) = prev.taskStarted(split, slot) } -class SplitRDD[T: ClassManifest](prev: RDD[T]) +class SplitRDD[T: ClassManifest](prev: RDD[T]) extends RDD[Array[T]](prev.sparkContext) { override def splits = prev.splits override def preferredLocations(split: Split) = prev.preferredLocations(split) @@ -170,16 +170,16 @@ extends RDD[Array[T]](prev.sparkContext) { } class SampledRDD[T: ClassManifest]( - prev: RDD[T], withReplacement: Boolean, frac: Double, seed: Int) + prev: RDD[T], withReplacement: Boolean, frac: Double, seed: Int) extends RDD[T](prev.sparkContext) { - + @transient val splits_ = { val rg = new Random(seed); prev.splits.map(x => new SeededSplit(x, rg.nextInt)) } override def splits = splits_.asInstanceOf[Array[Split]] override def preferredLocations(split: Split) = prev.preferredLocations(split.asInstanceOf[SeededSplit].prev) - override def iterator(splitIn: Split) = { + override def iterator(splitIn: Split) = { val split = splitIn.asInstanceOf[SeededSplit] val rg = new Random(split.seed); // Sampling with replacement (TODO: use reservoir sampling to make this more efficient?) @@ -213,7 +213,7 @@ extends RDD[T](prev.sparkContext) with Logging { else prev.preferredLocations(split) } - + override def iterator(split: Split): Iterator[T] = { val key = id + "::" + split.getId() logInfo("CachedRDD split key is " + key) @@ -278,7 +278,7 @@ extends Split { class UnionRDD[T: ClassManifest](sc: SparkContext, rdds: Seq[RDD[T]]) extends RDD[T](sc) { @transient val splits_ : Array[Split] = { - val splits: Seq[Split] = + val splits: Seq[Split] = for (rdd <- rdds; split <- rdd.splits) yield new UnionSplit(rdd, split) splits.toArray @@ -289,7 +289,7 @@ extends RDD[T](sc) { override def iterator(s: Split): Iterator[T] = s.asInstanceOf[UnionSplit[T]].iterator() - override def preferredLocations(s: Split): Seq[String] = + override def preferredLocations(s: Split): Seq[String] = s.asInstanceOf[UnionSplit[T]].preferredLocations() } -- GitLab