core/src/main/java/org/apache/hadoop/chukwa/datacollection/adaptor/filetailer/CharFileTailingAdaptorUTF8NewLineEscaped.java

/*
 * 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.hadoop.chukwa.datacollection.adaptor.filetailer;


import org.apache.hadoop.chukwa.ChunkImpl;
import org.apache.hadoop.chukwa.datacollection.ChunkReceiver;
import org.apache.hadoop.chukwa.util.RecordConstants;

import java.util.ArrayList;
import java.util.Arrays;

/**
 * A subclass of FileTailingAdaptor that reads UTF8/ascii files and splits
 * records at non-escaped carriage returns
 * 
 */
public class CharFileTailingAdaptorUTF8NewLineEscaped extends
    FileTailingAdaptor {

  private static final char SEPARATOR = '\n';

  private ArrayList<Integer> offsets = new ArrayList<Integer>();

  /**
   * 
   * Note: this method uses a temporary ArrayList (shared across instances).
   * This means we're copying ints each time. This could be a performance issue.
   * Also, 'offsets' never shrinks, and will be of size proportional to the
   * largest number of lines ever seen in an event.
   */
  @Override
  protected int extractRecords(ChunkReceiver eq, long buffOffsetInFile,
      byte[] buf) throws InterruptedException {
    String es = RecordConstants.RECORD_SEPARATOR_ESCAPE_SEQ;
    for (int i = 0; i < buf.length; ++i) {
      // if this is a separator
      if (buf[i] == SEPARATOR) {
        // if possibly preceded by escape sequence (avoid outOfBounds here)
        boolean escaped = false; // was it escaped?
        if (i - es.length() >= 0) {
          escaped = true; // maybe (at least there was room for an escape
                          // sequence, so let's check for the e.s.)
          for (int j = 0; j < es.length(); j++) {
            if (buf[i - es.length() + j] != es.charAt(j)) {
              escaped = false;
            }
          }
        }
        if (!escaped) {
          offsets.add(i);
        }
      }
    }

    if (offsets.size() > 0) {
      int[] offsets_i = new int[offsets.size()];
      for (int i = 0; i < offsets.size(); i++) {
        try {
        offsets_i[i] = offsets.get(i);
        } catch(NullPointerException e) {
          // Skip offsets 0 where it can be null.
        }
      }
      // make the stream unique to this adaptor
      int bytesUsed = 0;
      if(buf.length==offsets_i[offsets_i.length -1]) {
        // If Separator is last character of stream,
        // send the record.
        bytesUsed = offsets_i[offsets_i.length - 2] + 1;
      } else {
        // If the last record is partial read,
        // truncate the record to the n -1 new line.
        bytesUsed = offsets_i[offsets_i.length - 1] + 1; // char at last        
      }
                                                           // offset uses a byte
      assert bytesUsed > 0 : " shouldn't send empty events";
      ChunkImpl chunk = new ChunkImpl(type, toWatch.getAbsolutePath(),
          buffOffsetInFile + bytesUsed, Arrays.copyOf(buf, bytesUsed), this);

      chunk.setSeqID(buffOffsetInFile + bytesUsed);
      chunk.setRecordOffsets(offsets_i);
      eq.add(chunk);

      offsets.clear();
      return bytesUsed;
    } else
      return 0;
  }

  public String toString() {
    return "escaped newline CFTA-UTF8";
  }

}