* 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.

*/

* The storage driver for writing RCFile data through HowlOutputFormat.

*/

public class RCFileOutputDriver extends HowlOutputStorageDriver {

   /** The serde for serializing the HowlRecord to bytes writable */

   private SerDe serde;

   /** The object inspector for the given schema */

   private StructObjectInspector objectInspector;

   /** The schema for the output data */

   private HowlSchema outputSchema;

   /** The cached RCFile output format instance */

   private OutputFormat outputFormat = null;

  /* (non-Javadoc)

   * @see org.apache.hadoop.hive.howl.mapreduce.HowlOutputStorageDriver#convertValue(org.apache.hadoop.hive.howl.data.HowlRecord)

   */

  @Override

  public Writable convertValue(HowlRecord value) throws IOException {

    try {

      return serde.serialize(value.getAll(), objectInspector);

    } catch(SerDeException e) {

      throw new IOException(e);

    }

  }

  /* (non-Javadoc)

   * @see org.apache.hadoop.hive.howl.mapreduce.HowlOutputStorageDriver#generateKey(org.apache.hadoop.hive.howl.data.HowlRecord)

   */

  @Override

  public WritableComparable<?> generateKey(HowlRecord value) throws IOException {

    //key is not used for RCFile output

    return null;

  }

  /* (non-Javadoc)

   * @see org.apache.hadoop.hive.howl.mapreduce.HowlOutputStorageDriver#getOutputFormat(java.util.Properties)

   */

  @SuppressWarnings("unchecked")

  @Override

  public OutputFormat<? super WritableComparable<?>, ? super Writable> getOutputFormat() throws IOException {

    if( outputFormat == null ) {

      outputFormat = new RCFileMapReduceOutputFormat();

    }

    return outputFormat;

  }

  /* (non-Javadoc)

   * @see org.apache.hadoop.hive.howl.mapreduce.HowlOutputStorageDriver#setOutputPath(org.apache.hadoop.mapreduce.JobContext, java.lang.String)

   */

  @Override

  public void setOutputPath(JobContext jobContext, String location) throws IOException {

    //Not calling FileOutputFormat.setOutputPath since that requires a Job instead of JobContext

    jobContext.getConfiguration().set("mapred.output.dir", location);

  }

  /* (non-Javadoc)

   * @see org.apache.hadoop.hive.howl.mapreduce.HowlOutputStorageDriver#setPartitionValues(org.apache.hadoop.mapreduce.JobContext, java.util.Map)

   */

  @Override

  public void setPartitionValues(JobContext jobContext, Map<String, String> partitionValues)

      throws IOException {

    //default implementation of HowlOutputStorageDriver.getPartitionLocation will use the partition

    //values to generate the data location, so partition values not used here

  }

  /* (non-Javadoc)

   * @see org.apache.hadoop.hive.howl.mapreduce.HowlOutputStorageDriver#setSchema(org.apache.hadoop.mapreduce.JobContext, org.apache.hadoop.hive.metastore.api.Schema)

   */

  @Override

  public void setSchema(JobContext jobContext, HowlSchema schema) throws IOException {

    outputSchema = schema;

    RCFileMapReduceOutputFormat.setColumnNumber(

        jobContext.getConfiguration(), schema.getFields().size());

  }

  @Override

  public void initialize(JobContext context,Properties howlProperties) throws IOException {

    super.initialize(context, howlProperties);

    List<FieldSchema> fields = HowlUtil.getFieldSchemaList(outputSchema.getFields());

    howlProperties.setProperty(Constants.LIST_COLUMNS,

          MetaStoreUtils.getColumnNamesFromFieldSchema(fields));

    howlProperties.setProperty(Constants.LIST_COLUMN_TYPES,

          MetaStoreUtils.getColumnTypesFromFieldSchema(fields));

    // setting these props to match LazySimpleSerde

    howlProperties.setProperty(Constants.SERIALIZATION_NULL_FORMAT, "\\N");

    howlProperties.setProperty(Constants.SERIALIZATION_FORMAT, "1");

    try {

      serde = new ColumnarSerDe();

      serde.initialize(context.getConfiguration(), howlProperties);

      objectInspector = createStructObjectInspector();

    } catch (SerDeException e) {

      throw new IOException(e);

    }

  }

  public StructObjectInspector createStructObjectInspector() throws IOException {

    if( outputSchema == null ) {

      throw new IOException("Invalid output schema specified");

    }

    List<ObjectInspector> fieldInspectors = new ArrayList<ObjectInspector>();

    List<String> fieldNames = new ArrayList<String>();

    for(HowlFieldSchema howlFieldSchema : outputSchema.getFields()) {

      TypeInfo type = TypeInfoUtils.getTypeInfoFromTypeString(howlFieldSchema.getTypeString());

      fieldNames.add(howlFieldSchema.getName());

      fieldInspectors.add(getObjectInspector(type));

    }

    StructObjectInspector structInspector = ObjectInspectorFactory.

        getStandardStructObjectInspector(fieldNames, fieldInspectors);

    return structInspector;

  }

  public ObjectInspector getObjectInspector(TypeInfo type) throws IOException {

    switch(type.getCategory()) {

    case PRIMITIVE :

      PrimitiveTypeInfo primitiveType = (PrimitiveTypeInfo) type;

      return PrimitiveObjectInspectorFactory.

        getPrimitiveJavaObjectInspector(primitiveType.getPrimitiveCategory());

    case MAP :

      MapTypeInfo mapType = (MapTypeInfo) type;

      MapObjectInspector mapInspector = ObjectInspectorFactory.getStandardMapObjectInspector(

          getObjectInspector(mapType.getMapKeyTypeInfo()), getObjectInspector(mapType.getMapValueTypeInfo()));

      return mapInspector;

    case LIST :

      ListTypeInfo listType = (ListTypeInfo) type;

      ListObjectInspector listInspector = ObjectInspectorFactory.getStandardListObjectInspector(

          getObjectInspector(listType.getListElementTypeInfo()));

      return listInspector;

    case STRUCT :

      StructTypeInfo structType = (StructTypeInfo) type;

      List<TypeInfo> fieldTypes = structType.getAllStructFieldTypeInfos();

      List<ObjectInspector> fieldInspectors = new ArrayList<ObjectInspector>();

      for(TypeInfo fieldType : fieldTypes) {

        fieldInspectors.add(getObjectInspector(fieldType));

      }

      StructObjectInspector structInspector = ObjectInspectorFactory.getStandardStructObjectInspector(

          structType.getAllStructFieldNames(), fieldInspectors);

      return structInspector;

    default :

      throw new IOException("Unknown field schema type");

    }

  }

}