apex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (APEXMALHAR-2014) ParquetReader operator
Date Tue, 29 Mar 2016 12:43:25 GMT

    [ https://issues.apache.org/jira/browse/APEXMALHAR-2014?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15215952#comment-15215952
] 

ASF GitHub Bot commented on APEXMALHAR-2014:
--------------------------------------------

Github user shubham-pathak22 commented on a diff in the pull request:

    https://github.com/apache/incubator-apex-malhar/pull/219#discussion_r57716067
  
    --- Diff: contrib/src/main/java/com/datatorrent/contrib/parquet/ParquetFilePOJOReader.java
---
    @@ -0,0 +1,305 @@
    +/**
    + * 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 com.datatorrent.contrib.parquet;
    +
    +import java.lang.reflect.Field;
    +import java.util.ArrayList;
    +import java.util.List;
    +import java.util.Map;
    +import java.util.StringTokenizer;
    +
    +import org.apache.commons.lang3.ClassUtils;
    +
    +import com.google.common.collect.Lists;
    +import com.google.common.collect.Maps;
    +
    +import com.datatorrent.api.Context;
    +import com.datatorrent.api.DefaultOutputPort;
    +import com.datatorrent.api.annotation.OutputPortFieldAnnotation;
    +import com.datatorrent.lib.util.FieldInfo;
    +import com.datatorrent.lib.util.FieldInfo.SupportType;
    +import com.datatorrent.lib.util.PojoUtils;
    +import com.datatorrent.lib.util.PojoUtils.Setter;
    +
    +import parquet.example.data.Group;
    +import parquet.io.ParquetEncodingException;
    +
    +/**
    + * <p>
    + * ParquetFilePOJOReader
    + * </p>
    + * ParquetFilePOJOReader operator is a concrete implementation of
    + * AbstractParquetFileReader to read Parquet files and emit records as POJOs.The
    + * POJO class name & field mapping should be provided by the user. If this
    + * mapping is not provided then reflection is used to determine this mapping. As
    + * of now only primitive types ( INT32, INT64, BOOLEAN, FLOAT, DOUBLE, BINARY )
    + * are supported.
    + *
    + * @displayName ParquetFilePOJOReader
    + * @tags parquet,input adapter
    + * @since 3.3.0
    + */
    +public class ParquetFilePOJOReader extends AbstractParquetFileReader<Object>
    +{
    +
    +  /**
    +   * POJO class
    +   */
    +  protected transient Class<?> pojoClass;
    +  /**
    +   * Map containing setters for fields in POJO
    +   */
    +  protected transient Map<String, Setter> pojoSetters;
    +  /**
    +   * String representing Parquet TO POJO field mapping. If not provided, then
    +   * reflection is used to determine the mapping. Format :
    +   * PARQUET_FIELD_NAME:POJO_FIELD_NAME:TYPE
    +   * E.g.event_id:event_id_v2:INTEGER,org_id:org_id_v2:STRING,long_id:
    +   * long_id_v2:
    +   * LONG,css_file_loaded:css_file_loaded_v2:BOOLEAN,float_val:float_val_v2:
    +   * FLOAT,double_val:double_val_v2:DOUBLE
    +   */
    +  protected transient String groupToPOJOFieldsMapping = null;
    +  protected transient List<FieldInfo> fieldInfos;
    +  protected transient List<ActiveFieldInfo> columnFieldSetters = null;
    +  protected static final String FIELD_SEPARATOR = ":";
    +  protected static final String RECORD_SEPARATOR = ",";
    +
    +  @OutputPortFieldAnnotation(schemaRequired = true)
    +  public final transient DefaultOutputPort<Object> output = new DefaultOutputPort<Object>()
    +  {
    +
    +    @Override
    +    public void setup(Context.PortContext context)
    +    {
    +      pojoClass = context.getValue(Context.PortContext.TUPLE_CLASS);
    +      pojoSetters = Maps.newHashMap();
    +      for (Field f : pojoClass.getDeclaredFields()) {
    +        try {
    +          pojoSetters.put(f.getName(), generateSettersForField(pojoClass, f.getName()));
    +        } catch (NoSuchFieldException | SecurityException e) {
    +          e.printStackTrace();
    +        }
    +      }
    +
    +      if (groupToPOJOFieldsMapping == null) {
    +        fieldInfos = createFieldInfoMap(generateFieldInfoInputs());
    +      } else {
    +        fieldInfos = createFieldInfoMap(groupToPOJOFieldsMapping);
    +      }
    +      initColumnFieldSetters();
    +    }
    +
    +  };
    +
    +  /**
    +   * Converts Group to POJO
    +   */
    +  @Override
    +  protected Object convertGroup(Group group)
    +  {
    +    Object obj;
    +    try {
    +      obj = pojoClass.newInstance();
    +    } catch (InstantiationException | IllegalAccessException ex) {
    +      throw new RuntimeException(ex);
    +    }
    +    for (int i = 0; i < columnFieldSetters.size(); i++) {
    +      ParquetFilePOJOReader.ActiveFieldInfo afi = columnFieldSetters.get(i);
    +
    --- End diff --
    
    At present, Schema evolution in parquet supports adding new columns. In our case, if new
columns are added then user would need to update both, the parquet schema and pojo class.
Adding a try/catch does not help in that case. 
    However, I there might be a case where there are more fields in POJO as compared to schema
and that would throw InvalidRecordException. I have added try/catch to handle it.


> ParquetReader operator
> ----------------------
>
>                 Key: APEXMALHAR-2014
>                 URL: https://issues.apache.org/jira/browse/APEXMALHAR-2014
>             Project: Apache Apex Malhar
>          Issue Type: New Feature
>            Reporter: shubham pathak
>            Assignee: shubham pathak
>
> Developing a ParquetReaderOperator which would allow apex users to read records from
parquet files.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message