Fix bug in SizeTracking*

andrewor14 · andrewor14 · commit 2b7ee66c8a27 · 2014-07-18T11:44:44.000-07:00
We were not incorporating the delta in sizes, because Seq does not match the case a :: b :: c, while List does. This was originally reported by @mateiz.
diff --git a/core/src/main/scala/org/apache/spark/util/collection/SizeTracker.scala b/core/src/main/scala/org/apache/spark/util/collection/SizeTracker.scala
@@ -80,7 +80,7 @@ private[spark] trait SizeTracker {
     if (samples.size > 2) {
       samples.remove(0)
     }
-    val bytesDelta = samples.toSeq.reverse match {
+    val bytesDelta = samples.toList.reverse match {
       case latest :: previous :: tail =>
         (latest.size - previous.size).toDouble / (latest.numUpdates - previous.numUpdates)
       // If fewer than 2 samples, assume no change

Original file line number	Diff line number	Diff line change
`@@ -80,7 +80,7 @@ private[spark] trait SizeTracker {`
`80`	`80`	`if (samples.size > 2) {`
`81`	`81`	`samples.remove(0)`
`82`	`82`	`}`
`83`		`- val bytesDelta = samples.toSeq.reverse match {`
	`83`	`+ val bytesDelta = samples.toList.reverse match {`
`84`	`84`	`case latest :: previous :: tail =>`
`85`	`85`	`(latest.size - previous.size).toDouble / (latest.numUpdates - previous.numUpdates)`
`86`	`86`	`// If fewer than 2 samples, assume no change`