Shell Programming and Scripting

Hi Guys,

Once again I have a query regarding grouping the columns, below is my requirement:

Input File:
COL1; COL2; AMT1;AMT2
PORT1;CURR1;100;50
PORT1;CURR1;200;100
PORT2;CURR2;300;150
PORT3;CURR3;400;200
PORT3;CURR3;500;250

Expected Output:
COL1; COL2; AMT1;AMT2
PORT1;CURR1;300;150
PORT2;CURR2;300;150
PORT3;CURR3;900;450

How can I pass to values in below command:
/usr/bin/nawk 'BEGIN{FS=OFS=";"}NR==1{print;next}{a[$1";"$2]+=$3}END{for(i in a)print i, a[i]}' <INPUT FILE>

Please suggest so that I can get both AMT1 and AMT2 gouped by COL1 and COL2.
Thanks in advance.

Regards,
Sandeep

Code:

nawk '
BEGIN { FS=OFS=";" }
NR==1 { print ; next }
{
   id = $1 ";" $2;
   amt1[id] += $3;
   amt2[id] += $4;
}
END {
   for (id in amt1)
      print id, amt1[id], amt2[id];
}
' inputfile

Jean-Pierre.

Or with a little adjustment of my solution:

Code:

awk 'BEGIN{FS=OFS=";"}                              
NR==1{print;next}
{a[$1";"$2]+=$3; b[$1";"$2]+=$4}
END{for(i in a)print i, a[i], b[i]}' file

Thanks Aigles and Franklin.
I already got it work.

Regards,
Sandeep