Java Code Examples for org.apache.avro.mapreduce.AvroJob#setInputKeySchema()

The following examples show how to use org.apache.avro.mapreduce.AvroJob#setInputKeySchema() . You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may check out the related API usage on the sidebar.
Example 1
Source File: MRCompactorAvroKeyDedupJobRunner.java    From incubator-gobblin with Apache License 2.0 5 votes vote down vote up 
private void configureSchema(Job job) throws IOException {
  Schema newestSchema = getNewestSchemaFromSource(job, this.fs);
  if (this.useSingleInputSchema) {
    AvroJob.setInputKeySchema(job, newestSchema);
  }
  AvroJob.setMapOutputKeySchema(job, this.shouldDeduplicate ? getKeySchema(job, newestSchema) : newestSchema);
  AvroJob.setMapOutputValueSchema(job, newestSchema);
  AvroJob.setOutputKeySchema(job, newestSchema);
}
Example 2
Source File: CompactionAvroJobConfigurator.java    From incubator-gobblin with Apache License 2.0 5 votes vote down vote up 
@Override
protected void configureSchema(Job job) throws IOException {
  Schema newestSchema = MRCompactorAvroKeyDedupJobRunner.getNewestSchemaFromSource(job, this.fs);
  if (newestSchema != null) {
    if (this.state.getPropAsBoolean(MRCompactorAvroKeyDedupJobRunner.COMPACTION_JOB_AVRO_SINGLE_INPUT_SCHEMA, true)) {
      AvroJob.setInputKeySchema(job, newestSchema);
    }
    AvroJob.setMapOutputKeySchema(job, this.shouldDeduplicate ? getDedupKeySchema(newestSchema) : newestSchema);
    AvroJob.setMapOutputValueSchema(job, newestSchema);
    AvroJob.setOutputKeySchema(job, newestSchema);
  }
}
Example 3
Source File: OSMRunner.java    From geowave with Apache License 2.0 4 votes vote down vote up 
@Override
public int run(final String[] args) throws Exception {

  final Configuration conf = getConf();
  conf.set("tableName", ingestOptions.getQualifiedTableName());
  conf.set("osmVisibility", ingestOptions.getVisibilityOptions().getVisibility());

  // job settings
  final Job job = Job.getInstance(conf, ingestOptions.getJobName());
  job.setJarByClass(OSMRunner.class);

  switch (ingestOptions.getMapperType()) {
    case "NODE": {
      configureSchema(AvroNode.getClassSchema());
      inputAvroFile = ingestOptions.getNodesBasePath();
      job.setMapperClass(OSMNodeMapper.class);
      break;
    }
    case "WAY": {
      configureSchema(AvroWay.getClassSchema());
      inputAvroFile = ingestOptions.getWaysBasePath();
      job.setMapperClass(OSMWayMapper.class);
      break;
    }
    case "RELATION": {
      configureSchema(AvroRelation.getClassSchema());
      inputAvroFile = ingestOptions.getRelationsBasePath();
      job.setMapperClass(OSMRelationMapper.class);
      break;
    }
    default:
      break;
  }
  if ((avroSchema == null) || (inputAvroFile == null)) {
    throw new MissingArgumentException(
        "argument for mapper type must be one of: NODE, WAY, or RELATION");
  }

  enableLocalityGroups(ingestOptions);

  // input format
  job.setInputFormatClass(AvroKeyInputFormat.class);
  FileInputFormat.setInputPaths(job, inputAvroFile);
  AvroJob.setInputKeySchema(job, avroSchema);

  // mappper

  job.setOutputKeyClass(Text.class);
  job.setOutputValueClass(Mutation.class);
  job.setOutputFormatClass(AccumuloOutputFormat.class);
  AccumuloOutputFormat.setConnectorInfo(
      job,
      accumuloOptions.getUser(),
      new PasswordToken(accumuloOptions.getPassword()));
  AccumuloOutputFormat.setCreateTables(job, true);
  AccumuloOutputFormat.setDefaultTableName(job, ingestOptions.getQualifiedTableName());
  AccumuloOutputFormat.setZooKeeperInstance(
      job,
      new ClientConfiguration().withInstance(accumuloOptions.getInstance()).withZkHosts(
          accumuloOptions.getZookeeper()));

  // reducer
  job.setNumReduceTasks(0);

  return job.waitForCompletion(true) ? 0 : -1;
}