UNIX for Advanced & Expert Users

Don't you mean that NF > 0. Also I would change it slightly to be more robust
awk 'NR > 1 && NF == 3 { print $0 > $1".txt"}' this will also take care of usual garbage input lines

I Have a file

34EMEA|VODAFONE ES|5484|CPN|04/01/07|1|300
34LA|VODAFONE ES|5484|CPN|04/01/07|1|300

how can i aggregate it to
34EMEA|VODAFONE ES|5484|CPN|04/01/07|1|600

awk -F"|" '{ sum[$4] += $7 } END { for (name in sum) print name, sum[name] }' data_file

this aggregarates if column 4 is similar. but if columns 2 to 5 are similare i want to aggregate.
so when i tried
awk -F"|" '{ sum[$2]sum[$3]sum[4] += $7 } END { for (name in sum) print name, sum[name] }' data_file

it is not working. can u please help.
Many thanks

Code:

awk -F"|" '{ sum[$2 FS $3 FS $4] += $7 } END { for (name in sum) print name, sum[name] }' data_file

Vgersh,
WOW!!! Thats great and helped me..

but when i use this command,

awk -F"|" '{ sum[$2 FS $3 FS $4 FS $5] += $7 } END { for (name in sum) print name"|"sum[name] }' data_file

it prints only fields 2 to 7, i m not able to print first filed. I tried
...print $1"|"name"|"sum[name] }' data_file but is not working..
could you please help. Thanks a million..

Code:

sum[$1 FS $2 FS $3 FS $4 FS $5]

Hi Vgersh,

Thanks for your response. But the code

sum[$1 FS $2 FS $3 FS $4 FS $5] will compare fields from 1 to 5.

I need to comapre fields from 2 to 5 and aggregate, while printing i should print the first field also. The first field can be anyone within the matching lines (lines having similar value from 2nd to 5th coulmn)

23ASIA|VODAFONE ES|5484|000000000000314521|03/31/07|1|100
56NA|VODAFONE ES|5484|000000000000314521|03/31/07|1|200

then the output should be

either 23ASIA|VODAFONE ES|5484|000000000000314521|03/31/07|1|300
OR 56NA|VODAFONE ES|5484|000000000000314521|03/31/07|1|300

can we acheive this?

Code:

awk -F"|" '{ idx=$2 FS $3 FS $4 FS $5; sum[idx] += $7; first[idx]=$1 } END { for (name in sum) printf("%s%s%s%s%s\n", first[name], FS, name, FS, sum[name]) }' data_file