Hi Gurus,

In my file I have an amount field from position 74 to 87, which contains values starting with '+' as well as '-'. I want to add all positive values in a varible called "CREDIT" and all negative values in a variable "DEBIT". I know, we can use grep to identify values with positive and negative signs. But, don't know, how to add them in 2 separate variables in a script. Is there any way to do that on command line? Any help would be much appriciated. Thanks!

In my file Amount field has right justified values like,

+1345778500000
-327800000
+642307000000
-4300000
+2308007
These values look left justified here on the screen, but they are actually right justified in my file.

try something like that....

.. messed up the right justification...

Thanks jim mcnamara for the reply. It works fine, when values are left justified. But, I have right justified values in the file. So, '+' OR '-' can start at any position from 74 to 87 bytes.

How can we use 'length' function to treat the values as left justified, so the leading spaces can be ignored?

I think, following will solve the problem, where I don't care about position of '+' and '-' signs in the values. I can just check values > 0 and the rest, and according to that sum them in 2 different variables as below:

awk ' {
value=substr($0,74,14)
if(value>0) { plus+=value}
else {minus+=value}
}
END { print plus, minus} ' filename | read credit debit

Following works fine,

#!/bin/ksh

credit_sum=`awk '{ value=substr($0, 74, 14); if(value>0) {credit+=value}} END {print credit}' filename`

debit_sum=`awk '{ value=substr($0, 74, 14); if(value<0) {debit+=value}} END {print debit}' filename`

exit 0;