All Downloads are FREE. Search and download functionalities are using the official Maven repository.

org.apache.avro.mapred.DelegatingInputFormat Maven / Gradle / Ivy

/**
 * 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.avro.mapred;

import java.io.IOException;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.LinkedList;
import java.util.List;
import java.util.Map;
import java.util.Map.Entry;

import org.apache.avro.Schema;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.mapred.FileInputFormat;
import org.apache.hadoop.mapred.InputFormat;
import org.apache.hadoop.mapred.InputSplit;
import org.apache.hadoop.mapred.JobConf;
import org.apache.hadoop.mapred.RecordReader;
import org.apache.hadoop.mapred.Reporter;
import org.apache.hadoop.util.ReflectionUtils;

/**
 * An {@link InputFormat} that delegates read behavior of paths based on
 * their associated avro schema.
 * @see MultipleInputs#addInputPath(JobConf, Path, Class, Class)
 */
class DelegatingInputFormat implements InputFormat {

  public InputSplit[] getSplits(JobConf conf, int numSplits) throws IOException {

    JobConf confCopy = new JobConf(conf);
    List splits = new ArrayList();

    Map> mapperMap = AvroMultipleInputs
       .getMapperTypeMap(conf);
    Map schemaMap = AvroMultipleInputs
        .getInputSchemaMap(conf);
    Map> schemaPaths
        = new HashMap>();

    // First, build a map of Schemas to Paths
    for (Entry entry : schemaMap.entrySet()) {
      if (!schemaPaths.containsKey(entry.getValue())) {
        schemaPaths.put(entry.getValue(), new LinkedList());
        System.out.println(entry.getValue());
        System.out.println(entry.getKey());
      }

      schemaPaths.get(entry.getValue()).add(entry.getKey());
    }

    for (Entry> schemaEntry :
        schemaPaths.entrySet()) {
      Schema schema = schemaEntry.getKey();
      System.out.println(schema);
      InputFormat format = (InputFormat) ReflectionUtils.newInstance(
         AvroInputFormat.class, conf);
      List paths = schemaEntry.getValue();

      Map, List> mapperPaths
          = new HashMap, List>();

      // Now, for each set of paths that have a common Schema, build
      // a map of Mappers to the paths they're used for
      for (Path path : paths) {
       Class mapperClass = mapperMap.get(path);
       if (!mapperPaths.containsKey(mapperClass)) {
         mapperPaths.put(mapperClass, new LinkedList());
       }

       mapperPaths.get(mapperClass).add(path);
      }

      // Now each set of paths that has a common InputFormat and Mapper can
      // be added to the same job, and split together.
      for (Entry, List> mapEntry : mapperPaths
         .entrySet()) {
       paths = mapEntry.getValue();
       Class mapperClass = mapEntry.getKey();

       if (mapperClass == null) {
         mapperClass = (Class) conf.getMapperClass();
       }

       FileInputFormat.setInputPaths(confCopy, paths.toArray(new Path[paths
           .size()]));

       // Get splits for each input path and tag with InputFormat
       // and Mapper types by wrapping in a TaggedInputSplit.
       InputSplit[] pathSplits = format.getSplits(confCopy, numSplits);
       for (InputSplit pathSplit : pathSplits) {
         splits.add(new TaggedInputSplit(pathSplit, conf, format.getClass(),
             mapperClass, schema));
       }
      }
    }

    return splits.toArray(new InputSplit[splits.size()]);
  }

  @SuppressWarnings("unchecked")
  public RecordReader getRecordReader(InputSplit split, JobConf conf,
      Reporter reporter) throws IOException {

    // Find the Schema and then build the RecordReader from the
    // TaggedInputSplit.

    TaggedInputSplit taggedInputSplit = (TaggedInputSplit) split;
    Schema schema = taggedInputSplit.getSchema();
    AvroJob.setInputSchema(conf, schema);
    InputFormat inputFormat = (InputFormat) ReflectionUtils
       .newInstance(taggedInputSplit.getInputFormatClass(), conf);
    return inputFormat.getRecordReader(taggedInputSplit.getInputSplit(), conf,
       reporter);
  }
}




© 2015 - 2025 Weber Informatics LLC | Privacy Policy