Hi everybody... Please help me regarding this issue asap. we have requirement ... source file contains a column contains a id (primary key) and other(2nd) column which is having one or more values like (toys,car,bus) here comma is not a filed delimeter and . Filed delimiter is a '|'. for example Sample input: 100|toys,car,bus 200|hat,soap,kite,lux 300|jugs,mug Sample output should be like this: 100,toys 100,cars 100,bus 200,hat 200,soap 200,kite 200,lux 300,jugs 300,mug 2nd column may contain values dynamically...... 1st column is integer and 2nd column is longvarchar(2000).... please help me regarding this.... Thanks Prakash

I'm assuming you meant to say that after parsing the data into 2 columns by pipe symbol that the second column was then parsed by commas. Maybe the script below will do what you want. The only hard part is this: eval echo \"\$${i}\" As var i changes, this evaluates and prints the contents of ${i} Let me know if you have any questions. #!/bin/ksh i=2 while IFS="|" read col1 col2 do # parse column 2 by commas set - $(IFS=","; echo ${col2}) i=1 while (($i <= $#)) do printf "%d,%s\n" $col1 $(eval echo \"\$${i}\") ((i+=1)) done done < datafile