I need to write a shell script where I have to create a duplicate record based on a condition and then change the value in a column of the new row which will be created.

I am trying to create a duplicate record based on a condition and then replace on column in the duplicate row.


Input will be



If the last column ie; col5 is EUR, then one more row should be created with change in column value.

Output should be



3 Answers 3


Here are a couple of ways:

  • awk

    awk -F, '{print; if($NF=="EUR"){$NF="AUD"; OFS=",";print}}' file

    That is checking the last field ($NR) to check another field, change $NR to $N, where N is the number you want to check (e.g $3 for the 5th field).

  • Perl

    perl -pe '$_.= "$1AUD\n" if /(.*)EUR$/' file 

    This one uses a lot of Perl shorthand that is kind of harder to explain. You can write the same thing like this:

    perl -e 'while(<>){if(/(.*)EUR$/){print "$_","$1AUD\n"} else {print "$_"}}' 

for each line of the input file

  • The perl one-liner will print the new record (with AUD) before the old one (with EUR)
    – Joseph R.
    Commented Sep 23, 2013 at 13:41
  • Thanks Joseph, Just to add, How can I change the position of condition column, FOr example if the condition column is Col3 instead of Col5.
    – Rakesh K
    Commented Sep 23, 2013 at 14:05
  • @user47766 see updated answer, the gawk one is easy to change, the Perl on is a bit more complex.
    – terdon
    Commented Sep 23, 2013 at 17:43
  • @user47766 Are we still talking about substituting EUR with AUD but a in a different column or will the patterns be different in this case?
    – Joseph R.
    Commented Sep 23, 2013 at 18:13

using sed:

 sed -e 's|^\(.*\),EUR$|\1,EUR\r\1,AUD|'

A Perl alternative

perl -ne 'print;print if s/EUR\Z/AUD/' your_file

If you want to make the modifications to the file in place, just add the -i switch:

perl -ni -e 'print;print if s/EUR\Z/AUD/' your_file

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .