Hi ,
The above problem you can achieve by using reverse pivot transform.
In the first query you can take one extra filed(New_filed) with data type integer and do the mapping like "GEN_ROW_NUM_BY_GROUP(COLUMN1)
In the next step you can take the reverse pivot transform do the below configurations:
Non-pivot columns: Column1
pivot columns : colimn2 and default value put null
pivot axis column : New_filed
axis values and prefix value
1 & col3
2 & col4
3 & col5
after that you can validate the job and you can see the fields in the schema output.
In the next query-- Take the output filed Coulmn1
and take the new output filed and the mapping like below:
rtrim((col3||','||col4||','||col5),',')
after that you pass this fileds into target table.