[jira] [Updated] (SQOOP-2037) Sqoop2: HDFSConnector need to send BigDecimal for decimal type and JODA for date type in Object array

[Veena Basavaraj (JIRA)]

     [ 
https://issues.apache.org/jira/browse/SQOOP-2037?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Veena Basavaraj updated SQOOP-2037:
-----------------------------------
    Description: 
At this point, we do not see the HDFS Connector sending BigDecimal for objects 
that are Sqoop Decimal type, as we prescribe JODA for date objects, we need 
BigDecimal for decimal objects.

writeSting is fine, but writeArray needs to according sqoop object format.
code to be fixed
{code}

    rowsRead++;
      if (HdfsUtils.hasCustomFormat(linkConfiguration, fromJobConfiguration)) {
        dataWriter.writeArrayRecord(HdfsUtils.formatRecord(linkConfiguration, 
fromJobConfiguration, line.toString()));
      } else {
        dataWriter.writeStringRecord(line.toString());
      }
    }

{code}


  was:
At this point, we do not see the Kite Connector sending BigDecimal for objects 
that are Sqoop Decimal type, as we prescribe JODA for date objects, we need 
BigDecimal for decimal objects.

code to be fixed
{code}

  @Override
  public void extract(ExtractorContext context, LinkConfiguration linkConfig,
      FromJobConfiguration fromJobConfig, KiteDatasetPartition partition) {
    String uri = partition.getUri();
    LOG.info("Loading data from " + uri);

    KiteDatasetExecutor executor = getExecutor(uri);
    DataWriter writer = context.getDataWriter();
    Object[] array;
    rowsRead = 0L;

    try {
      while ((array = executor.readRecord()) != null) {
        // TODO: SQOOP-1616 will cover more column data types. Use schema and 
do data type conversion (e.g. datatime).
        writer.writeArrayRecord(array);
        rowsRead++;
      }
    } finally {
      executor.closeReader();
    }
  }

{code}



> Sqoop2: HDFSConnector need to send BigDecimal for decimal type and JODA for 
> date type in Object array
> -----------------------------------------------------------------------------------------------------
>
>                 Key: SQOOP-2037
>                 URL: https://issues.apache.org/jira/browse/SQOOP-2037
>             Project: Sqoop
>          Issue Type: Bug
>            Reporter: Veena Basavaraj
>            Assignee: Qian Xu
>             Fix For: 1.99.5
>
>
> At this point, we do not see the HDFS Connector sending BigDecimal for 
> objects that are Sqoop Decimal type, as we prescribe JODA for date objects, 
> we need BigDecimal for decimal objects.
> writeSting is fine, but writeArray needs to according sqoop object format.
> code to be fixed
> {code}
>     rowsRead++;
>       if (HdfsUtils.hasCustomFormat(linkConfiguration, fromJobConfiguration)) 
> {
>         dataWriter.writeArrayRecord(HdfsUtils.formatRecord(linkConfiguration, 
> fromJobConfiguration, line.toString()));
>       } else {
>         dataWriter.writeStringRecord(line.toString());
>       }
>     }
> {code}



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