SPARK-3020: Print completed indices rather than tasks in web UI

Author: Patrick Wendell <[email protected]>

Closes #1933 from pwendell/speculation and squashes the following commits:

33a3473 [Patrick Wendell] Use OpenHashSet
8ce2ff0 [Patrick Wendell] SPARK-3020: Print completed indices rather than tasks in web UI

Author: pwendell

core/src/main/scala/org/apache/spark/ui/jobs/JobProgressListener.scala

@@ -153,6 +153,7 @@ class JobProgressListener(conf: SparkConf) extends SparkListener with Logging {
       val (errorMessage, metrics): (Option[String], Option[TaskMetrics]) =
         taskEnd.reason match {
           case org.apache.spark.Success =>
+            stageData.completedIndices.add(info.index)
             stageData.numCompleteTasks += 1
             (None, Option(taskEnd.taskMetrics))
           case e: ExceptionFailure =>  // Handle ExceptionFailure because we might have metrics

core/src/main/scala/org/apache/spark/ui/jobs/StageTable.scala

@@ -168,7 +168,7 @@ private[ui] class StageTableBase(
     <td valign="middle">{submissionTime}</td>
     <td sorttable_customkey={duration.getOrElse(-1).toString}>{formattedDuration}</td>
     <td class="progress-cell">
-      {makeProgressBar(stageData.numActiveTasks, stageData.numCompleteTasks,
+      {makeProgressBar(stageData.numActiveTasks, stageData.completedIndices.size,
         stageData.numFailedTasks, s.numTasks)}
     </td>
     <td sorttable_customekey={inputRead.toString}>{inputReadWithUnit}</td>

core/src/main/scala/org/apache/spark/ui/jobs/UIData.scala

@@ -19,6 +19,7 @@ package org.apache.spark.ui.jobs
 
 import org.apache.spark.executor.TaskMetrics
 import org.apache.spark.scheduler.{AccumulableInfo, TaskInfo}
+import org.apache.spark.util.collection.OpenHashSet
 
 import scala.collection.mutable.HashMap
 
@@ -38,6 +39,7 @@ private[jobs] object UIData {
   class StageUIData {
     var numActiveTasks: Int = _
     var numCompleteTasks: Int = _
+    var completedIndices = new OpenHashSet[Int]()
     var numFailedTasks: Int = _
 
     var executorRunTime: Long = _