Changes between Initial Version and Version 1 of waue/2009/0728


Ignore:
Timestamp:
Jul 28, 2009, 5:38:17 PM (15 years ago)
Author:
waue
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • waue/2009/0728

    v1 v1  
     1{{{
     2
     3/**
     4 * Copyright 2008 The Apache Software Foundation
     5 *
     6 * Licensed to the Apache Software Foundation (ASF) under one
     7 * or more contributor license agreements.  See the NOTICE file
     8 * distributed with this work for additional information
     9 * regarding copyright ownership.  The ASF licenses this file
     10 * to you under the Apache License, Version 2.0 (the
     11 * "License"); you may not use this file except in compliance
     12 * with the License.  You may obtain a copy of the License at
     13 *
     14 *     http://www.apache.org/licenses/LICENSE-2.0
     15 *
     16 * Unless required by applicable law or agreed to in writing, software
     17 * distributed under the License is distributed on an "AS IS" BASIS,
     18 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     19 * See the License for the specific language governing permissions and
     20 * limitations under the License.
     21 */
     22package hbase;
     23
     24import java.io.IOException;
     25import java.util.Map;
     26
     27import org.apache.hadoop.conf.Configuration;
     28import org.apache.hadoop.fs.Path;
     29import org.apache.hadoop.hbase.HBaseConfiguration;
     30import org.apache.hadoop.hbase.io.Cell;
     31import org.apache.hadoop.hbase.io.HbaseMapWritable;
     32import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
     33import org.apache.hadoop.hbase.io.RowResult;
     34import org.apache.hadoop.hbase.util.Bytes;
     35import org.apache.hadoop.mapred.FileOutputFormat;
     36import org.apache.hadoop.mapred.JobClient;
     37import org.apache.hadoop.mapred.JobConf;
     38import org.apache.hadoop.mapred.OutputCollector;
     39import org.apache.hadoop.mapred.Reporter;
     40import org.apache.hadoop.mapred.lib.IdentityReducer;
     41import org.apache.hadoop.util.Tool;
     42import org.apache.hadoop.util.ToolRunner;
     43
     44/**
     45 * A job with a map to count rows.
     46 * Map outputs table rows IF the input row has columns that have content. 
     47 * Uses an {@link IdentityReducer}
     48 */
     49public class RowCounter extends TableMap<ImmutableBytesWritable, RowResult> implements Tool {
     50  /* Name of this 'program'
     51   */
     52  static final String NAME = "rowcounter";
     53 
     54  private Configuration conf;
     55  private final RowResult EMPTY_RESULT_VALUE =
     56        new RowResult(Bytes.toBytes("dummy"),new HbaseMapWritable<byte [], Cell>());
     57  private static enum Counters {ROWS}
     58 
     59  @Override
     60  public void map(ImmutableBytesWritable row, RowResult value,
     61    OutputCollector<ImmutableBytesWritable, RowResult> output,
     62    @SuppressWarnings("unused") Reporter reporter)
     63  throws IOException {
     64    boolean content = false;
     65    for (Map.Entry<byte [], Cell> e: value.entrySet()) {
     66      Cell cell = e.getValue();
     67      if (cell != null && cell.getValue().length > 0) {
     68        content = true;
     69        break;
     70      }
     71    }
     72    if (!content) {
     73      return;
     74    }
     75    // Give out same value every time.  We're only interested in the row/key
     76    reporter.incrCounter(Counters.ROWS, 1);
     77    output.collect(row, EMPTY_RESULT_VALUE);
     78  }
     79
     80  /**
     81   * @param args
     82   * @return the JobConf
     83   * @throws IOException
     84   */
     85  @SuppressWarnings({ "unused", "deprecation" })
     86  public JobConf createSubmittableJob(String[] args) throws IOException {
     87    JobConf c = new JobConf(getConf(), RowCounter.class);
     88    c.setJobName(NAME);
     89    // Columns are space delimited
     90    StringBuilder sb = new StringBuilder();
     91    final int columnoffset = 2;
     92    for (int i = columnoffset; i < args.length; i++) {
     93      if (i > columnoffset) {
     94        sb.append(" ");
     95      }
     96      sb.append(args[i]);
     97    }
     98    // Second argument is the table name.
     99    TableMap.initJob(args[1], sb.toString(), this.getClass(),
     100      ImmutableBytesWritable.class, RowResult.class, c);
     101    c.setReducerClass(IdentityReducer.class);
     102    // First arg is the output directory.
     103    FileOutputFormat.setOutputPath(c, new Path(args[0]));
     104    return c;
     105  }
     106 
     107  static int printUsage() {
     108    System.out.println(NAME +
     109      " <outputdir> <tablename> <column1> [<column2>...]");
     110    return -1;
     111  }
     112 
     113  public int run(final String[] args) throws Exception {
     114    // Make sure there are at least 3 parameters
     115    if (args.length < 3) {
     116      System.err.println("ERROR: Wrong number of parameters: " + args.length);
     117      return printUsage();
     118    }
     119    JobClient.runJob(createSubmittableJob(args));
     120    return 0;
     121  }
     122
     123  public Configuration getConf() {
     124    return this.conf;
     125  }
     126
     127  public void setConf(final Configuration c) {
     128    this.conf = c;
     129  }
     130
     131  /**
     132   * @param args
     133   * @throws Exception
     134   */
     135  public static void main(String[] args) throws Exception {
     136         String[] argv = {"/user/waue/hba","t1","f2:c1"};
     137         args = argv;
     138    HBaseConfiguration c = new HBaseConfiguration();
     139    int errCode = ToolRunner.run(c, new RowCounter(), args);
     140    System.exit(errCode);
     141  }
     142}
     143
     144}}}