mjhjmtu

I am trying to aggregate a file containing the following data in UNIX.
I need to add the amounts if the key is same.

Key,amount,date,Time
abc-xyz-12234,45,15-08-91,23:00
pqr-vgh-5241,15,15-08-91,21:00
abc-xyz-12234,35,15-08-91,23:00
pqr-vgh-5241,24,15-08-91,21:00
abc-xyz-12234,655,15-08-91,23:00
lkj-erf-8542,281,15-08-91,10:00
pqr-vgh-5241,40,15-08-91,21:00

Output should be as following

abc-xyz-12234,735,15-08-91,23:00
pqr-vgh-5241,79,15-08-91,21:00
lkj-erf-8542,281,15-08-91,10:00

I tried by the following command ,but it just gives me uniq

cat file | grep "abc-xyz-12234" | uniq

Another possible solution with awk could be:

awk 'BEGIN FS = OFS = "," 
 NR != 1 y[$1] += $2; $2 = y[$1]; x[$1] = $0; 
 END for (i in x) print x[i]; ' file

The flaw is it won't preserve your order. So result could be:

pqr-vgh-5241,79,15-08-91,21:00
abc-xyz-12234,735,15-08-91,23:00
lkj-erf-8542,281,15-08-91,10:00

You can do this with awk:

#!/bin/sh
sort | awk -F, '
function result() 
 if ( key != "" ) 
 printf "%s,%d,%sn", key, value, datetime;
 

BEGIN key = ""; value = 0; datetime = ""; 
$2 ~ /^[0-9]+/ 
 if ( $1 == key ) 
 value += $2;
 else 
 result();
 key = $1;
 value = $2;
 datetime = $3 "," $4;
 

END result(); 
'

giving

./foo <input
abc-xyz-12234,735,15-08-91,23:00
lkj-erf-8542,281,15-08-91,10:00
pqr-vgh-5241,79,15-08-91,21:00

Here's a way in Perl. Call it as such ./script file.ext:

use warnings;
use strict;

my %data;
my @order;

while (<>)
 next if $. == 1;
 my @line = split /,/;
 if (defined $data$line[0])
 $data$line[0]->[1] += $line[1];
 
 else 
 $data$line[0] = @line;
 push @order, $line[0];
 


for (@order)
 print join(',', @$data$_);