Reverting again everthing to 'Rebase on trunk' commit as my local branch was dirty

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/branches/MR-2454@1415014 13f79535-47bb-0310-9956-ffa450edef68

hadoop-mapreduce-project/CHANGES.txt

@@ -1,8 +1,3 @@
-Branch MR-2454
-
-   MAPREDUCE-4809. Make internal classes required for MAPREDUCE-2454 to be
-   java public. (masokan via tucu)
-
 Hadoop MapReduce Change Log
 
 Trunk (Unreleased)

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/MapTask.java

@@ -34,8 +34,6 @@ import java.util.concurrent.locks.ReentrantLock;
 
 import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
-import org.apache.hadoop.classification.InterfaceAudience;
-import org.apache.hadoop.classification.InterfaceStability;
 import org.apache.hadoop.fs.FSDataInputStream;
 import org.apache.hadoop.fs.FSDataOutputStream;
 import org.apache.hadoop.fs.FileSystem;
@@ -56,7 +54,6 @@ import org.apache.hadoop.io.serializer.Serializer;
 import org.apache.hadoop.mapred.IFile.Writer;
 import org.apache.hadoop.mapred.Merger.Segment;
 import org.apache.hadoop.mapred.SortedRanges.SkipRangeIterator;
-import org.apache.hadoop.mapreduce.JobContext;
 import org.apache.hadoop.mapreduce.MRJobConfig;
 import org.apache.hadoop.mapreduce.TaskAttemptContext;
 import org.apache.hadoop.mapreduce.TaskCounter;
@@ -74,9 +71,7 @@ import org.apache.hadoop.util.StringInterner;
 import org.apache.hadoop.util.StringUtils;
 
 /** A Map task. */
-@InterfaceAudience.LimitedPrivate({"MapReduce"})
-@InterfaceStability.Unstable
-public class MapTask extends Task {
+class MapTask extends Task {
   /**
    * The size of each record in the index file for the map-outputs.
    */
@@ -343,10 +338,6 @@ public class MapTask extends Task {
     done(umbilical, reporter);
   }
 
-  public Progress getSortPhase() {
-    return sortPhase;
-  }
-
  @SuppressWarnings("unchecked")
  private <T> T getSplitDetails(Path file, long offset) 
   throws IOException {
@@ -375,22 +366,6 @@ public class MapTask extends Task {
    return split;
  }
   
-  @SuppressWarnings("unchecked")
-  private <KEY, VALUE> MapOutputCollector<KEY, VALUE>
-          createSortingCollector(JobConf job, TaskReporter reporter)
-    throws IOException, ClassNotFoundException {
-    MapOutputCollector<KEY, VALUE> collector
-      = (MapOutputCollector<KEY, VALUE>)
-       ReflectionUtils.newInstance(
-                        job.getClass(JobContext.MAP_OUTPUT_COLLECTOR_CLASS_ATTR,
-                        MapOutputBuffer.class, MapOutputCollector.class), job);
-    LOG.info("Map output collector class = " + collector.getClass().getName());
-    MapOutputCollector.Context context =
-                           new MapOutputCollector.Context(this, job, reporter);
-    collector.init(context);
-    return collector;
-  }
-
   @SuppressWarnings("unchecked")
   private <INKEY,INVALUE,OUTKEY,OUTVALUE>
   void runOldMapper(final JobConf job,
@@ -413,14 +388,11 @@ public class MapTask extends Task {
 
     int numReduceTasks = conf.getNumReduceTasks();
     LOG.info("numReduceTasks: " + numReduceTasks);
-    MapOutputCollector<OUTKEY, OUTVALUE> collector = null;
+    MapOutputCollector collector = null;
     if (numReduceTasks > 0) {
-      collector = createSortingCollector(job, reporter);
+      collector = new MapOutputBuffer(umbilical, job, reporter);
     } else { 
-      collector = new DirectMapOutputCollector<OUTKEY, OUTVALUE>();
-       MapOutputCollector.Context context =
-                           new MapOutputCollector.Context(this, job, reporter);
-      collector.init(context);
+      collector = new DirectMapOutputCollector(umbilical, job, reporter);
     }
     MapRunnable<INKEY,INVALUE,OUTKEY,OUTVALUE> runner =
       ReflectionUtils.newInstance(job.getMapRunnerClass(), job);
@@ -666,7 +638,7 @@ public class MapTask extends Task {
                        TaskUmbilicalProtocol umbilical,
                        TaskReporter reporter
                        ) throws IOException, ClassNotFoundException {
-      collector = createSortingCollector(job, reporter);
+      collector = new MapOutputBuffer<K,V>(umbilical, job, reporter);
       partitions = jobContext.getNumReduceTasks();
       if (partitions > 1) {
         partitioner = (org.apache.hadoop.mapreduce.Partitioner<K,V>)
@@ -762,6 +734,17 @@ public class MapTask extends Task {
     output.close(mapperContext);
   }
 
+  interface MapOutputCollector<K, V> {
+
+    public void collect(K key, V value, int partition
+                        ) throws IOException, InterruptedException;
+    public void close() throws IOException, InterruptedException;
+    
+    public void flush() throws IOException, InterruptedException, 
+                               ClassNotFoundException;
+        
+  }
+
   class DirectMapOutputCollector<K, V>
     implements MapOutputCollector<K, V> {
  
@@ -769,18 +752,14 @@ public class MapTask extends Task {
 
     private TaskReporter reporter = null;
 
-    private Counters.Counter mapOutputRecordCounter;
-    private Counters.Counter fileOutputByteCounter;
-    private List<Statistics> fsStats;
-
-    public DirectMapOutputCollector() {
-    }
+    private final Counters.Counter mapOutputRecordCounter;
+    private final Counters.Counter fileOutputByteCounter;
+    private final List<Statistics> fsStats;
 
     @SuppressWarnings("unchecked")
-    public void init(MapOutputCollector.Context context
-                    ) throws IOException, ClassNotFoundException {
-      this.reporter = context.getReporter();
-      JobConf job = context.getJobConf();
+    public DirectMapOutputCollector(TaskUmbilicalProtocol umbilical,
+        JobConf job, TaskReporter reporter) throws IOException {
+      this.reporter = reporter;
       String finalName = getOutputName(getPartition());
       FileSystem fs = FileSystem.get(job);
 
@@ -837,23 +816,23 @@ public class MapTask extends Task {
 
   private class MapOutputBuffer<K extends Object, V extends Object>
       implements MapOutputCollector<K, V>, IndexedSortable {
-    private int partitions;
-    private JobConf job;
-    private TaskReporter reporter;
-    private Class<K> keyClass;
-    private Class<V> valClass;
-    private RawComparator<K> comparator;
-    private SerializationFactory serializationFactory;
-    private Serializer<K> keySerializer;
-    private Serializer<V> valSerializer;
-    private CombinerRunner<K,V> combinerRunner;
-    private CombineOutputCollector<K, V> combineCollector;
+    final int partitions;
+    final JobConf job;
+    final TaskReporter reporter;
+    final Class<K> keyClass;
+    final Class<V> valClass;
+    final RawComparator<K> comparator;
+    final SerializationFactory serializationFactory;
+    final Serializer<K> keySerializer;
+    final Serializer<V> valSerializer;
+    final CombinerRunner<K,V> combinerRunner;
+    final CombineOutputCollector<K, V> combineCollector;
 
     // Compression for map-outputs
-    private CompressionCodec codec;
+    final CompressionCodec codec;
 
     // k/v accounting
-    private IntBuffer kvmeta; // metadata overlay on backing store
+    final IntBuffer kvmeta; // metadata overlay on backing store
     int kvstart;            // marks origin of spill metadata
     int kvend;              // marks end of spill metadata
     int kvindex;            // marks end of fully serialized records
@@ -877,15 +856,15 @@ public class MapTask extends Task {
     private static final int METASIZE = NMETA * 4; // size in bytes
 
     // spill accounting
-    private int maxRec;
-    private int softLimit;
+    final int maxRec;
+    final int softLimit;
     boolean spillInProgress;;
     int bufferRemaining;
     volatile Throwable sortSpillException = null;
 
     int numSpills = 0;
-    private int minSpillsForCombine;
-    private IndexedSorter sorter;
+    final int minSpillsForCombine;
+    final IndexedSorter sorter;
     final ReentrantLock spillLock = new ReentrantLock();
     final Condition spillDone = spillLock.newCondition();
     final Condition spillReady = spillLock.newCondition();
@@ -893,12 +872,12 @@ public class MapTask extends Task {
     volatile boolean spillThreadRunning = false;
     final SpillThread spillThread = new SpillThread();
 
-    private FileSystem rfs;
+    final FileSystem rfs;
 
     // Counters
-    private Counters.Counter mapOutputByteCounter;
-    private Counters.Counter mapOutputRecordCounter;
-    private Counters.Counter fileOutputByteCounter;
+    final Counters.Counter mapOutputByteCounter;
+    final Counters.Counter mapOutputRecordCounter;
+    final Counters.Counter fileOutputByteCounter;
 
     final ArrayList<SpillRecord> indexCacheList =
       new ArrayList<SpillRecord>();
@@ -906,23 +885,12 @@ public class MapTask extends Task {
     private int indexCacheMemoryLimit;
     private static final int INDEX_CACHE_MEMORY_LIMIT_DEFAULT = 1024 * 1024;
 
-    private MapTask mapTask;
-    private MapOutputFile mapOutputFile;
-    private Progress sortPhase;
-    private Counters.Counter spilledRecordsCounter;
-
-    public MapOutputBuffer() {
-    }
-
     @SuppressWarnings("unchecked")
-    public void init(MapOutputCollector.Context context
-                    ) throws IOException, ClassNotFoundException {
-      job = context.getJobConf();
-      reporter = context.getReporter();
-      mapTask = context.getMapTask();
-      mapOutputFile = mapTask.getMapOutputFile();
-      sortPhase = mapTask.getSortPhase();
-      spilledRecordsCounter = reporter.getCounter(TaskCounter.SPILLED_RECORDS);
+    public MapOutputBuffer(TaskUmbilicalProtocol umbilical, JobConf job,
+                           TaskReporter reporter
+                           ) throws IOException, ClassNotFoundException {
+      this.job = job;
+      this.reporter = reporter;
       partitions = job.getNumReduceTasks();
       rfs = ((LocalFileSystem)FileSystem.getLocal(job)).getRaw();
 
@@ -999,7 +967,7 @@ public class MapTask extends Task {
       if (combinerRunner != null) {
         final Counters.Counter combineOutputCounter =
           reporter.getCounter(TaskCounter.COMBINE_OUTPUT_RECORDS);
-        combineCollector= new CombineOutputCollector<K,V>(combineOutputCounter, reporter, job);
+        combineCollector= new CombineOutputCollector<K,V>(combineOutputCounter, reporter, conf);
       } else {
         combineCollector = null;
       }
@@ -1150,10 +1118,6 @@ public class MapTask extends Task {
       }
     }
 
-    private TaskAttemptID getTaskID() {
-      return mapTask.getTaskID();
-    }
-
     /**
      * Set the point from which meta and serialization data expand. The meta
      * indices are aligned with the buffer, so metadata never spans the ends of
@@ -1526,7 +1490,7 @@ public class MapTask extends Task {
         if (lspillException instanceof Error) {
           final String logMsg = "Task " + getTaskID() + " failed : " +
             StringUtils.stringifyException(lspillException);
-          mapTask.reportFatalError(getTaskID(), lspillException, logMsg);
+          reportFatalError(getTaskID(), lspillException, logMsg);
         }
         throw new IOException("Spill failed", lspillException);
       }

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/SpillRecord.java

@@ -34,7 +34,7 @@ import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.io.IOUtils;
 import org.apache.hadoop.util.PureJavaCrc32;
 
-public class SpillRecord {
+class SpillRecord {
 
   /** Backing store */
   private final ByteBuffer buf;
@@ -143,3 +143,17 @@ public class SpillRecord {
   }
 
 }
+
+class IndexRecord {
+  long startOffset;
+  long rawLength;
+  long partLength;
+
+  public IndexRecord() { }
+
+  public IndexRecord(long startOffset, long rawLength, long partLength) {
+    this.startOffset = startOffset;
+    this.rawLength = rawLength;
+    this.partLength = partLength;
+  }
+}

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/Task.java

@@ -584,9 +584,9 @@ abstract public class Task implements Writable, Configurable {
     return status;
   }
 
-  @InterfaceAudience.LimitedPrivate({"MapReduce"})
+  @InterfaceAudience.Private
   @InterfaceStability.Unstable
-  public class TaskReporter 
+  protected class TaskReporter 
       extends org.apache.hadoop.mapreduce.StatusReporter
       implements Runnable, Reporter {
     private TaskUmbilicalProtocol umbilical;
@@ -1466,9 +1466,9 @@ abstract public class Task implements Writable, Configurable {
     return reducerContext;
   }
 
-  @InterfaceAudience.LimitedPrivate({"MapReduce"})
+  @InterfaceAudience.Private
   @InterfaceStability.Unstable
-  public static abstract class CombinerRunner<K,V> {
+  protected static abstract class CombinerRunner<K,V> {
     protected final Counters.Counter inputCounter;
     protected final JobConf job;
     protected final TaskReporter reporter;
@@ -1486,13 +1486,13 @@ abstract public class Task implements Writable, Configurable {
      * @param iterator the key/value pairs to use as input
      * @param collector the output collector
      */
-    public abstract void combine(RawKeyValueIterator iterator, 
+    abstract void combine(RawKeyValueIterator iterator, 
                           OutputCollector<K,V> collector
                          ) throws IOException, InterruptedException, 
                                   ClassNotFoundException;
 
     @SuppressWarnings("unchecked")
-    public static <K,V> 
+    static <K,V> 
     CombinerRunner<K,V> create(JobConf job,
                                TaskAttemptID taskId,
                                Counters.Counter inputCounter,
@@ -1542,7 +1542,7 @@ abstract public class Task implements Writable, Configurable {
     }
 
     @SuppressWarnings("unchecked")
-    public void combine(RawKeyValueIterator kvIter,
+    protected void combine(RawKeyValueIterator kvIter,
                            OutputCollector<K,V> combineCollector
                            ) throws IOException {
       Reducer<K,V,K,V> combiner = 
@@ -1611,7 +1611,7 @@ abstract public class Task implements Writable, Configurable {
 
     @SuppressWarnings("unchecked")
     @Override
-    public void combine(RawKeyValueIterator iterator, 
+    void combine(RawKeyValueIterator iterator, 
                  OutputCollector<K,V> collector
                  ) throws IOException, InterruptedException,
                           ClassNotFoundException {

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/MRJobConfig.java

@@ -30,9 +30,6 @@ public interface MRJobConfig {
 
   public static final String MAP_CLASS_ATTR = "mapreduce.job.map.class";
 
-  public static final String MAP_OUTPUT_COLLECTOR_CLASS_ATTR
-                                  = "mapreduce.job.map.output.collector.class";
-
   public static final String COMBINE_CLASS_ATTR = "mapreduce.job.combine.class";
 
   public static final String REDUCE_CLASS_ATTR = "mapreduce.job.reduce.class";

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/ExceptionReporter.java

@@ -17,14 +17,9 @@
  */
 package org.apache.hadoop.mapreduce.task.reduce;
 
-import org.apache.hadoop.classification.InterfaceAudience;
-import org.apache.hadoop.classification.InterfaceStability;
-
 /**
  * An interface for reporting exceptions to other threads
  */
-@InterfaceAudience.LimitedPrivate({"MapReduce"})
-@InterfaceStability.Unstable
-public interface ExceptionReporter {
+interface ExceptionReporter {
   void reportException(Throwable t);
 }

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/MapHost.java

@@ -20,14 +20,9 @@ package org.apache.hadoop.mapreduce.task.reduce;
 import java.util.ArrayList;
 import java.util.List;
 
-import org.apache.hadoop.classification.InterfaceAudience;
-import org.apache.hadoop.classification.InterfaceStability;
-
 import org.apache.hadoop.mapreduce.TaskAttemptID;
 
-@InterfaceAudience.LimitedPrivate({"MapReduce"})
-@InterfaceStability.Unstable
-public class MapHost {
+class MapHost {
   
   public static enum State {
     IDLE,               // No map outputs available

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/MapOutput.java

@@ -24,8 +24,6 @@ import java.util.concurrent.atomic.AtomicInteger;
 
 import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
-import org.apache.hadoop.classification.InterfaceAudience;
-import org.apache.hadoop.classification.InterfaceStability;
 import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.LocalDirAllocator;
@@ -35,9 +33,7 @@ import org.apache.hadoop.mapred.JobConf;
 import org.apache.hadoop.mapred.MapOutputFile;
 import org.apache.hadoop.mapreduce.TaskAttemptID;
 
-@InterfaceAudience.LimitedPrivate({"MapReduce"})
-@InterfaceStability.Unstable
-public class MapOutput<K,V> {
+class MapOutput<K,V> {
   private static final Log LOG = LogFactory.getLog(MapOutput.class);
   private static AtomicInteger ID = new AtomicInteger(0);
   
@@ -66,7 +62,7 @@ public class MapOutput<K,V> {
   
   private final boolean primaryMapOutput;
   
-  public MapOutput(TaskAttemptID mapId, MergeManager<K,V> merger, long size, 
+  MapOutput(TaskAttemptID mapId, MergeManager<K,V> merger, long size, 
             JobConf conf, LocalDirAllocator localDirAllocator,
             int fetcher, boolean primaryMapOutput, MapOutputFile mapOutputFile)
          throws IOException {
@@ -91,7 +87,7 @@ public class MapOutput<K,V> {
     this.primaryMapOutput = primaryMapOutput;
   }
   
-  public MapOutput(TaskAttemptID mapId, MergeManager<K,V> merger, int size, 
+  MapOutput(TaskAttemptID mapId, MergeManager<K,V> merger, int size, 
             boolean primaryMapOutput) {
     this.id = ID.incrementAndGet();
     this.mapId = mapId;

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/MergeManager.java

@@ -59,7 +59,7 @@ import org.apache.hadoop.util.Progress;
 import org.apache.hadoop.util.ReflectionUtils;
 
 @SuppressWarnings(value={"unchecked", "deprecation"})
-@InterfaceAudience.LimitedPrivate({"MapReduce"})
+@InterfaceAudience.Private
 @InterfaceStability.Unstable
 public class MergeManager<K, V> {
   

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/Shuffle.java

@@ -38,7 +38,7 @@ import org.apache.hadoop.mapreduce.MRJobConfig;
 import org.apache.hadoop.mapreduce.TaskAttemptID;
 import org.apache.hadoop.util.Progress;
 
-@InterfaceAudience.LimitedPrivate({"MapReduce"})
+@InterfaceAudience.Private
 @InterfaceStability.Unstable
 @SuppressWarnings({"unchecked", "rawtypes"})
 public class Shuffle<K, V> implements ExceptionReporter {

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/ShuffleClientMetrics.java

@@ -17,9 +17,6 @@
  */
 package org.apache.hadoop.mapreduce.task.reduce;
 
-import org.apache.hadoop.classification.InterfaceAudience;
-import org.apache.hadoop.classification.InterfaceStability;
-
 import org.apache.hadoop.mapred.JobConf;
 import org.apache.hadoop.mapreduce.MRJobConfig;
 import org.apache.hadoop.mapreduce.TaskAttemptID;
@@ -28,9 +25,7 @@ import org.apache.hadoop.metrics.MetricsRecord;
 import org.apache.hadoop.metrics.MetricsUtil;
 import org.apache.hadoop.metrics.Updater;
 
-@InterfaceAudience.LimitedPrivate({"MapReduce"})
-@InterfaceStability.Unstable
-public class ShuffleClientMetrics implements Updater {
+class ShuffleClientMetrics implements Updater {
 
   private MetricsRecord shuffleMetrics = null;
   private int numFailedFetches = 0;

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/resources/mapred-default.xml

@@ -928,12 +928,4 @@
   <value>jhs/_HOST@REALM.TLD</value>
 </property>
 
-<property>
-  <name>mapreduce.job.map.output.collector.class</name>
-  <value>org.apache.hadoop.mapred.MapTask$MapOutputBuffer</value>
-  <description>
-    It defines the MapOutputCollector implementation to use.
-  </description>
-</property>
-
 </configuration>