001/*
002 * Licensed to the Apache Software Foundation (ASF) under one
003 * or more contributor license agreements.  See the NOTICE file
004 * distributed with this work for additional information
005 * regarding copyright ownership.  The ASF licenses this file
006 * to you under the Apache License, Version 2.0 (the
007 * "License"); you may not use this file except in compliance
008 * with the License.  You may obtain a copy of the License at
009 *
010 *     http://www.apache.org/licenses/LICENSE-2.0
011 *
012 * Unless required by applicable law or agreed to in writing, software
013 * distributed under the License is distributed on an "AS IS" BASIS,
014 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
015 * See the License for the specific language governing permissions and
016 * limitations under the License.
017 */
018package org.apache.hadoop.hbase.mapreduce;
019
020import java.io.IOException;
021import java.nio.charset.Charset;
022import java.util.List;
023import org.apache.hadoop.hbase.client.TableDescriptor;
024import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
025import org.apache.hadoop.hbase.util.Bytes;
026import org.apache.hadoop.mapreduce.Job;
027import org.apache.yetus.audience.InterfaceAudience;
028import org.slf4j.Logger;
029import org.slf4j.LoggerFactory;
030
031/**
032 * Create 3 level tree directory, first level is using table name as parent directory and then use
033 * family name as child directory, and all related HFiles for one family are under child directory
034 * -tableName1 -columnFamilyName1 -columnFamilyName2 -HFiles -tableName2 -columnFamilyName1 -HFiles
035 * -columnFamilyName2
036 */
037@InterfaceAudience.Public
038public class MultiTableHFileOutputFormat extends HFileOutputFormat2 {
039  private static final Logger LOG = LoggerFactory.getLogger(MultiTableHFileOutputFormat.class);
040
041  /**
042   * Creates a composite key to use as a mapper output key when using
043   * MultiTableHFileOutputFormat.configureIncrementaLoad to set up bulk ingest job
044   * @param tableName Name of the Table - Eg: TableName.getNameAsString()
045   * @param suffix    Usually represents a rowkey when creating a mapper key or column family
046   * @return byte[] representation of composite key
047   */
048  public static byte[] createCompositeKey(byte[] tableName, byte[] suffix) {
049    return combineTableNameSuffix(tableName, suffix);
050  }
051
052  /**
053   * Alternate api which accepts an ImmutableBytesWritable for the suffix
054   * @see MultiTableHFileOutputFormat#createCompositeKey(byte[], byte[])
055   */
056  public static byte[] createCompositeKey(byte[] tableName, ImmutableBytesWritable suffix) {
057    return combineTableNameSuffix(tableName, suffix.get());
058  }
059
060  /**
061   * Alternate api which accepts a String for the tableName and ImmutableBytesWritable for the
062   * suffix
063   * @see MultiTableHFileOutputFormat#createCompositeKey(byte[], byte[])
064   */
065  public static byte[] createCompositeKey(String tableName, ImmutableBytesWritable suffix) {
066    return combineTableNameSuffix(tableName.getBytes(Charset.forName("UTF-8")), suffix.get());
067  }
068
069  /**
070   * Analogous to
071   * {@link HFileOutputFormat2#configureIncrementalLoad(Job, TableDescriptor, RegionLocator)}, this
072   * function will configure the requisite number of reducers to write HFiles for multple tables
073   * simultaneously
074   * @param job                   See {@link org.apache.hadoop.mapreduce.Job}
075   * @param multiTableDescriptors Table descriptor and region locator pairs n
076   */
077  public static void configureIncrementalLoad(Job job, List<TableInfo> multiTableDescriptors)
078    throws IOException {
079    MultiTableHFileOutputFormat.configureIncrementalLoad(job, multiTableDescriptors,
080      MultiTableHFileOutputFormat.class);
081  }
082
083  final private static int validateCompositeKey(byte[] keyBytes) {
084
085    int separatorIdx = Bytes.indexOf(keyBytes, tableSeparator);
086
087    // Either the separator was not found or a tablename wasn't present or a key wasn't present
088    if (separatorIdx == -1) {
089      throw new IllegalArgumentException("Invalid format for composite key ["
090        + Bytes.toStringBinary(keyBytes) + "]. Cannot extract tablename and suffix from key");
091    }
092    return separatorIdx;
093  }
094
095  protected static byte[] getTableName(byte[] keyBytes) {
096    int separatorIdx = validateCompositeKey(keyBytes);
097    return Bytes.copy(keyBytes, 0, separatorIdx);
098  }
099
100  protected static byte[] getSuffix(byte[] keyBytes) {
101    int separatorIdx = validateCompositeKey(keyBytes);
102    return Bytes.copy(keyBytes, separatorIdx + 1, keyBytes.length - separatorIdx - 1);
103  }
104}