Option F

Learn how to use the -F option in awk to customize the field separator when processing structured text. By default, awk splits records on whitespace, but -F lets you define any character or regular expression as the delimiter.

usage: awk [-F fs] [-v var=value] [-f progfile | 'prog'] [file ...]

1. Default Field Splitting

By default, awk treats any whitespace as the field separator:

$ df -h

Filesystem      Size  Used Avail Use% Mounted on
udev            1.8G     0  1.8G   0% /dev
tmpfs           366M  9.7M  356M   3% /run
...

When your data uses a different delimiter—such as commas, colons, or pipes—you must override this behavior.

2. Changing the Field Separator with -F

To use a colon (:) as the separator:

$ awk -F ":" '{ print $1 }' sizeV1.txt

Filesystem
udev
tmpfs
/dev/sda1
...

Print the second field:

$ awk -F ":" '{ print $2 }' sizeV1.txt

Size
1.8G
366M
...

Combine fields 1 and 5:

$ awk -F ":" '{ print $1, $5 }' sizeV1.txt

Filesystem Use%
udev 0%
tmpfs 3%
...

You can also pipe data into awk:

$ cat sizeV1.txt | awk -F ":" '{ print $2 }'

3. Why Quote or Escape the Separator

Certain characters (for example, |, &, *, <, >) have special meanings to the shell.

awk -F | '{ print $1 }' employees.txt

awk -F "|" '{ print $1 }' employees.txt
# or
awk -F \| '{ print $1 }' employees.txt

4. Common Field Separators

5. Example: Processing Database Output

Here’s a Bash script (db.sh) that runs a PostgreSQL query inside Docker, trimming each field:

#!/usr/bin/env bash
docker_run() {
  docker exec -i employees_db psql -U postgres -d employees -tAQ <<EOF
SELECT
  trim(id_employee::text),
  trim(first_name),
  trim(last_name),
  trim(area),
  trim(job_title),
  trim(email),
  trim(salary::text)
FROM "employee";
EOF
}
docker_run > employees.txt
exit 0

View the output:

$ cat employees.txt

1|Kriti|Shreshtha|Finance|Financial Analyst|[email protected]|60000
...

Extract first names (field 2):

$ awk -F "|" '{ print $2 }' employees.txt

Kriti
Rajasekar
Debbie
...

Next Steps

Next, we’ll explore the -v option to declare variables within awk:

usage: awk [-F fs] [-v var=value] [-f progfile | 'prog'] [file ...]

References

• GNU Awk Manual
• awk(1) — Linux manual page
• PostgreSQL Documentation