| /* |
| * Licensed to the Apache Software Foundation (ASF) under one |
| * or more contributor license agreements. See the NOTICE file |
| * distributed with this work for additional information |
| * regarding copyright ownership. The ASF licenses this file |
| * to you under the Apache License, Version 2.0 (the |
| * "License"); you may not use this file except in compliance |
| * with the License. You may obtain a copy of the License at |
| * |
| * http://www.apache.org/licenses/LICENSE-2.0 |
| * |
| * Unless required by applicable law or agreed to in writing, software |
| * distributed under the License is distributed on an "AS IS" BASIS, |
| * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| * See the License for the specific language governing permissions and |
| * limitations under the License. |
| */ |
| package org.apache.pig.impl.io; |
| |
| import java.io.DataInputStream; |
| import java.io.IOException; |
| |
| import org.apache.commons.collections4.queue.CircularFifoQueue; |
| import org.apache.hadoop.conf.Configuration; |
| import org.apache.hadoop.fs.FSDataInputStream; |
| import org.apache.hadoop.fs.FileSystem; |
| import org.apache.hadoop.fs.Path; |
| import org.apache.hadoop.io.Text; |
| import org.apache.hadoop.mapreduce.InputSplit; |
| import org.apache.hadoop.mapreduce.RecordReader; |
| import org.apache.hadoop.mapreduce.TaskAttemptContext; |
| import org.apache.hadoop.mapreduce.lib.input.FileSplit; |
| import org.apache.pig.backend.executionengine.ExecException; |
| import org.apache.pig.data.BinInterSedes; |
| import org.apache.pig.data.InterSedes; |
| import org.apache.pig.data.InterSedesFactory; |
| import org.apache.pig.data.Tuple; |
| |
| /** |
| * A record reader used to read data written using {@link InterRecordWriter} |
| * It uses the default InterSedes object for deserialization. |
| */ |
| public class InterRecordReader extends RecordReader<Text, Tuple> { |
| |
| private long start; |
| private long lastDataPos; |
| private long end; |
| private BufferedPositionedInputStream in; |
| private Tuple value = null; |
| private DataInputStream inData = null; |
| private static InterSedes sedes = InterSedesFactory.getInterSedesInstance(); |
| |
| private byte[] syncMarker; |
| private long lastSyncPos = -1; |
| private long syncMarkerInterval; |
| private long dataBytesSeen = 0; |
| |
| public InterRecordReader(int syncMarkerLength, long syncMarkerInterval) { |
| this.syncMarker = new byte[syncMarkerLength]; |
| this.syncMarkerInterval = syncMarkerInterval; |
| } |
| |
| public void initialize(InputSplit genericSplit, |
| TaskAttemptContext context) throws IOException { |
| FileSplit split = (FileSplit) genericSplit; |
| Configuration job = context.getConfiguration(); |
| start = split.getStart(); |
| end = start + split.getLength(); |
| final Path file = split.getPath(); |
| |
| // open the file |
| FileSystem fs = file.getFileSystem(job); |
| FSDataInputStream fileIn = fs.open(split.getPath()); |
| |
| // read the magic byte sequence serving as record marker but only if the file is not empty |
| if (!(start == 0 && end == 0)) { |
| fileIn.readFully(0, syncMarker, 0, syncMarker.length); |
| } |
| |
| //seek to the start of the split |
| fileIn.seek(start); |
| |
| in = new BufferedPositionedInputStream(fileIn, start); |
| inData = new DataInputStream(in); |
| } |
| |
| |
| /** |
| * Skips to next sync marker |
| * @return true if marker was observed, false if EOF or EndOfSplit was reached |
| * @throws IOException |
| */ |
| public boolean skipUntilMarkerOrSplitEndOrEOF() throws IOException { |
| int b = Integer.MIN_VALUE; |
| CircularFifoQueue<Integer> queue = new CircularFifoQueue(syncMarker.length); |
| outer:while (b != -1) { |
| //There may be a case where we read through a whole split without a marker, then we shouldn't proceed |
| // because the records are from the next split which another reader would pick up too |
| //One exception of reading past split end is if at least the first byte of the marker was seen before split |
| // end. |
| if (in.getPosition() >= (end+syncMarker.length-1)) { |
| return false; |
| } |
| b = in.read(); |
| |
| //EOF reached |
| if (b == -1) return false; |
| |
| queue.add(b); |
| if (queue.size() != queue.maxSize()) { |
| //Not enough bytes read yet |
| continue outer; |
| } |
| int i = 0; |
| for (Integer seenByte : queue){ |
| if (syncMarker[i++] != seenByte.byteValue()) { |
| continue outer; |
| } |
| } |
| //Found marker: queue content equals sync marker |
| lastSyncPos = in.getPosition(); |
| return true; |
| } |
| return false; |
| } |
| |
| /** |
| * Reads a sync marker |
| * @return true if sync marker was read, false if EOF reached |
| * @throws IOException thrown if neither EOF nor proper sync was found |
| */ |
| private boolean readSyncFullyOrEOF() throws IOException { |
| int b = in.read(); |
| if (b == -1) { |
| //EOF reached |
| return false; |
| } |
| if ((byte) b != syncMarker[0]) { |
| throw new IOException("Corrupt data file, expected sync marker at position " + in.getPosition()); |
| } |
| int i = 1; |
| while (i < syncMarker.length) { |
| b = in.read(); |
| if ((byte) b != syncMarker[i]) { |
| throw new IOException("Corrupt data file, expected sync marker at position " + in.getPosition()); |
| } |
| ++i; |
| } |
| lastSyncPos = in.getPosition(); |
| return true; |
| |
| } |
| |
| private boolean readDataOrEOF() throws IOException { |
| long preDataPos = in.getPosition(); |
| int b = in.read(); |
| if(!BinInterSedes.isTupleByte((byte) b) ) { |
| if (b == -1) { |
| //EOF reached |
| return false; |
| } else { |
| throw new IOException("Corrupt data file, expected tuple type byte, but seen " + b); |
| } |
| } |
| try { |
| value = (Tuple)sedes.readDatum(inData, (byte)b); |
| lastDataPos = in.getPosition(); |
| dataBytesSeen += (lastDataPos-preDataPos); |
| return true; |
| } catch (ExecException ee) { |
| throw ee; |
| } |
| } |
| |
| public boolean nextKeyValue() throws IOException { |
| |
| //No marker has been seen, look for next marker |
| if (lastSyncPos == -1) { |
| if (!skipUntilMarkerOrSplitEndOrEOF()) { |
| return false; |
| } |
| } |
| |
| //If we've read more or equal amount of data than the sync interval, we expect a sync marker or EOF |
| if (dataBytesSeen >= syncMarkerInterval) { |
| boolean isEOF = !readSyncFullyOrEOF(); |
| if (isEOF) { |
| return false; |
| } |
| dataBytesSeen = 0; |
| //If we've just seen a (non-first) sync marker which was completely in the next split then we need to stop |
| if (in.getPosition()-syncMarker.length >= end) { |
| return false; |
| } |
| } |
| |
| //Sync marker has been seen, expect data |
| return readDataOrEOF(); |
| } |
| |
| @Override |
| public Text getCurrentKey() { |
| // the key is always null since we don't really have a key for each |
| // input record |
| return null; |
| } |
| |
| @Override |
| public Tuple getCurrentValue() { |
| return value; |
| } |
| |
| /** |
| * Get the progress within the split |
| */ |
| public float getProgress() { |
| if (start == end) { |
| return 0.0f; |
| } else { |
| return Math.min(1.0f, (lastDataPos - start) / (float)(end - start)); |
| } |
| } |
| |
| public synchronized void close() throws IOException { |
| if (in != null) { |
| in.close(); |
| } |
| } |
| } |