Get values from the next row and merge- awk

Question

I have a pipe delimited file like this

    OLD|123432
    NEW|232322
    OLD|1234452
    NEW|232324
    OLD|656966
    NEW|232325

I am trying to create a new file where I am trying to merge rows based on the value in the first column (OLD/NEW). First column in the output file will have the new number and the second column will have the old number.

Output

232322|123432
232324|1234452
232325|656966

I looked at the answer here How to merge every two lines into one from the command line?. I know it is not the exact solution but used as a starting point.

and tried to make it work to solve this but throws syntax error.

awk -F "|" 'NR%2{OFS = "|" printf "%s ",$0;next;}1'

A `sed` one-liner could be `sed 'N;s/.*|$.*$\n.*|$.*$/\2|\1/' file` — M. Nejat Aydin, Feb 25 '21 at 20:58

score 2 · Answer 1 · answered Feb 25 '21 at 19:39

2

You may use this awk:

awk 'BEGIN {FS=OFS="|"} $1 == "NEW" {print $2, old} $1 == "OLD" {old = $2}' file

232322|123432
232324|1234452
232325|656966

answered Feb 25 '21 at 19:39

anubhava

761,203
64
569
643

score 1 · Accepted Answer · answered Feb 25 '21 at 21:26

Using $0 will have the value of the whole line. If the field separator is a pipe, the second column is $2 that has the number.

If you want to use the remainder with NR%2, another option could be storing the value of the second column in a variable, for example v

awk 'BEGIN{FS=OFS="|"} NR%2{v=$2;next;}{print $2,v}' file

Output

232322|123432
232324|1234452
232325|656966

Get values from the next row and merge- awk

2 Answers2