Help reg to write gawk script

Question

Need sample gawk to join two records with unique key $2.$1 has different entity eventhough $2 is key thats why there are ther are two different lines for same key

I/p FIle not in sorted order

Output file:

O/p file 01|12345|DDDDD 01|99|DDDDD 01|888||DDDDD

This is quite unclear. If the key is unique, why are there multiple records with that key? What exactly is the transformation rule you want to apply? — Anko, Apr 08 '15 at 21:16

score 1 · Answer 1 · answered Apr 08 '15 at 21:38

1

You can do that with awk, e.g. like this:

awk '
    BEGIN { FS=OFS="|" }
    $2!=old { print sav ; sav=$0 ; old=$2 ; next }
    { sub(/[^|]*\|[^|]*\|/,"") ; sav=sav OFS $0 }
    END { print sav }
'

answered Apr 08 '15 at 21:38

Janis

14,222

Thanks this help, added to the first request if the input is not sorted and expecting the o/p in diff way I/p FIle 01|12345|AAAA 01|99|AAAA 01|888|AAAA 02|12345|BBBBB|CCCCC|DDDDD 02|99|BBBBB|CCCCC|DDDDD 02|888|BBBBB|CCCCC|DDDDD
O/p file 01|12345|DDDDD 01|99|DDDDD 01|888|DDDDD
– dileep Apr 09 '15 at 18:53

Help reg to write gawk script

1 Answers1