How to print selected columns with awk

Reports and command exports often contain more fields than a handoff, ticket, or quick review needs. Printing only selected awk fields keeps the result tied to column positions instead of copying whole rows and trimming them by eye.

With -F '|', awk splits each row on pipe characters and stores the fields as $1, $2, $3, and later numbered columns. Setting OFS controls the separator inserted between printed fields when the print list uses commas.

The sample uses simple pipe-delimited rows with no quoted delimiters or multiline fields. For CSV exports with quoted commas or embedded line breaks, use a CSV-aware parser first, then pass normalized fields to awk.

1. Save a small pipe-delimited sample file.

   servers.txt

   server|role|region|status|owner
   api-01|web|us-east|ready|ops
   worker-02|queue|eu-west|drain|data
   cache-03|cache|us-east|ready|ops

2. Print only the columns needed for the review.

   $ awk -F'|' 'BEGIN{OFS=","}{print $1,$3,$4}' servers.txt
   server,region,status
   api-01,us-east,ready
   worker-02,eu-west,drain
   cache-03,us-east,ready

   -F'|' reads pipe-delimited input. print $1,$3,$4 selects the server, region, and status fields, and OFS writes commas between the printed fields.

3. Change the field order by changing the print list.

   $ awk -F'|' 'BEGIN{OFS=","}{print $4,$1}' servers.txt
   status,server
   ready,api-01
   drain,worker-02
   ready,cache-03

   The original input is unchanged. Only the selected fields and their output order change.

4. Replace the sample field separator and field numbers with the values from the real input.

   Use awk -F ',' for simple comma-delimited rows, awk -F '\t' for tab-delimited rows, or another separator that matches the source format.

5. Remove the sample file after testing.

   $ rm servers.txt