column Command Cheat Sheet

column formats its input into multiple columns. It is incredibly useful for turning raw, delimited text output (like CSVs or log files) into human-readable tables in the terminal.

Synopsis

column [options] [file...]

Description

The column utility formats its input into multiple columns. Rows are filled before columns. Input is taken from file operands, or, by default, from standard input.

Basic Usage

Auto-Format Text Table

cat data.txt | column -t

Specify Delimiter (-s)

Format a CSV file readable:

column -s "," -t file.csv

Fill Rows First (-x)

By default, column fills columns first (vertical list). Use -x to fill rows first (horizontal list).

ls -1 | column -x

Table Formatting Options (Modern Versions)

Newer versions of column (util-linux) support advanced table formatting.

Custom Output Separator (-o)

Change the separator between columns in the output (default is two spaces).

mount | column -t -o " | "

Hide Headers

If input has headers you want to clean up manually or re-process.

column --table-noheadings -t data.txt

JSON Output (-J)

Export table data as JSON. Requires specifying column names.

column -t -J --table-columns id,name,age data.txt

Define Column Names (-N)

If input lacks headers, you can provide them.

ps | column -t -N PID,TTY,TIME,CMD

Practical Examples

Format /etc/passwd

Make the password file readable by splitting on :.

column -s ":" -t /etc/passwd | head -n 5

Pretty Print Mount Points

mount | column -t

Align dmesg Output

dmesg | tail | column -t

Display CSV File as Table

# cat users.csv
# id,name,email
# 1,alice,alice@example.com
# 2,bob,bob@example.com

column -s, -t < users.csv

Output:

id  name   email
1   alice  alice@example.com
2   bob    bob@example.com

Monitor Disk Usage

Automate df -h alignment (though df usually aligns itself well, this ensures custom parsers work).

df -h | column -t

Advanced Usage

Right Align Specific Columns (-R)

Right-align the 2nd and 3rd columns (useful for numbers).

column -t -R 2,3 data.txt

Truncate Long Text (-T)

Truncate text in specific columns if it exceeds width.

column -t -T 4 very_long_log.txt

Tree-Like Output

If data represents a hierarchy (requires parent-child ID columns):

column --tree-id 1 --tree-parent 2 --tree 3 data.txt

Colors

Colorize output (if supported by terminal and column version).

column --table-columns USER,PID,%CPU,%MEM,COMMAND --color=always

Handling Pipeline Output

Often used with awk, sed, or grep to clean data before formatting.

grep "Error" /var/log/syslog | awk '{print $1,$2,$3,$5}' | column -t

Notes

• Debian/Ubuntu: column is part of bsdmainutils or util-linux.
• macOS (BSD): Older version of column comes pre-installed. It supports -t and -s but lacks JSON (-J) or advanced table options (--table-*).
• Whitespace: Default separator includes tabs and spaces. Use -s $'\t' for tab-only.

Compatibility Warning

The advanced flags (--table-*, -J, -N) are available in util-linux version (standard on most Linux distros like Ubuntu, Fedora, CentOS). BSD variants (macOS) generally only support basic -t, -s, and -x.

Check version:

column --version

Exit Status