Implement UDF and Pig Script logic in Talend

Question

I have written a UDF that reads an input file and segregates the Data into String and Integer or String and Double.

My UDF is working fine. Also I have written a Pig Script to use the above jar on HDFS.

Now I want to have this code integrated with Talend for Big Data. How can I acheive this.

The java code in the UDF is below :

package com.test.udf;


import java.io.IOException;
import org.apache.pig.EvalFunc;
import org.apache.pig.data.Tuple;
import org.apache.pig.data.TupleFactory;

public class CheckDataType extends EvalFunc<Tuple> {

       @Override
       public Tuple exec(Tuple input) throws IOException {
              // TODO Auto-generated method stub

              String valString = null;
              Integer valInt =  null;
              Double valDouble =null;
              String str = (String) input.get(0);

              Tuple outputTuple =TupleFactory.getInstance().newTuple(2);

              if (str != null){
                     try{
                           valInt = Integer.parseInt(str);
                           outputTuple.set(0, valString);
                           outputTuple.set(1, valInt);

                     }
                     catch(Exception e){
                           try{
                                  valDouble = Double.parseDouble(str) ;
                                  outputTuple.set(0, valString);
                                  outputTuple.set(1, valDouble);
                           }
                           catch(Exception ew){
                           outputTuple.set(0, str);
                           outputTuple.set(1, null);
                           }
                     }     

              }
              return outputTuple;

       }

}

Also The pig script I have written is below :

REGISTER 'CONVERT.jar';

data_load = LOAD '/tmp/input/testfile.txt' USING PigStorage(',') AS (col1:chararray, col2:chararray, col3:chararray, col4:chararray, col5:chararray);

data_grp = GROUP data_load BY ($input_col);

data_flatten = FOREACH data_grp GENERATE FLATTEN(com.test.udf.CheckDataType(*));

rmf /tmp/output;

STORE data_flatten INTO '/tmp/output' USING PigStorage(',');

How can I integrate this in Talend for Big Data.

Answer 1

Updated answer:

You need to split your pig script into 3 components, PigLoad, PigCode and PigStoreResult and connect them. UDF can be included as a code or as a seperate jar included to PigLoad component.

Step by step instruction can be found here: https://www.evernote.com/l/AJONeXS0_sBNwpDfmPByJSUVS0vmAs04EGM