column — Tabular Formatting

Format delimited data into aligned columns — the final presentation step before human eyes hit the terminal.

Tabulate Output — -t

Auto-align whitespace-separated output into columns

mount | column -t

Tabulate a colon-delimited file — specify separator with -s

column -t -s':' /etc/passwd

Tabulate comma-separated data

column -t -s',' data.csv

Tabulate pipe-delimited data

echo -e "name|age|city\nalice|30|portland\nbob|25|seattle" | column -t -s'|'

Prettify key-value output from any command

sysctl -a 2>/dev/null | head -20 | column -t -s'='

Input separator — tells column how to split fields

column -t -s$'\t' data.tsv

Output separator — -o flag sets the string between output columns

column -t -s':' -o'  |  ' /etc/passwd | head -5

Multiple input separators — use any of these characters to split

column -t -s':,' mixed_delimiters.txt

Convert tabular data to JSON

echo -e "name age city\nalice 30 portland\nbob 25 seattle" | column -t -J -N name,age,city

JSON output from a CSV with named columns

column -t -s',' -J -N hostname,ip,role < servers.csv

Name columns for JSON output or documentation

df -h | column -t -N filesystem,size,used,avail,pct,mount

Right-align specific columns — -R (column numbers, 1-indexed)

echo -e "service count bytes\nnginx 1500 45000\nsshd 200 8000" | column -t -R 2,3

Hide columns — -H (column numbers to suppress)

column -t -s':' -H 2 /etc/passwd | head -5
# Hides column 2 (password field)

Combine named columns with right-alignment

echo -e "name score grade\nalice 98 A\nbob 75 C" | column -t -N name,score,grade -R 2

Make mount output readable

mount | column -t

Format ss output with aligned columns

ss -tln | column -t

Align environment variables

env | sort | column -t -s'='

Pretty-print a CSV for quick inspection

head -20 report.csv | column -t -s','

Format key-value pairs from a config file

grep -v '^#' /etc/sysctl.conf | grep '=' | column -t -s'='

Align process list with specific fields

ps -eo pid,user,%cpu,%mem,comm --sort=-%cpu | head -15 | column -t

column: auto-aligns existing output — no format string needed

# Best when you already have the data and just need alignment
cat data.txt | column -t

printf/awk: precise control over width and formatting

# Best when you need exact column widths or numeric formatting
awk '{printf "%-20s %10d %8.2f\n", $1, $2, $3}' data.txt

column works on full output — awk works per-line during processing

# column needs all lines to compute max widths — buffers entire input
# awk printf uses fixed widths — streams, no buffering

Quick CSV viewer — header + separator + data

(head -1 data.csv; echo "---"; tail -n +2 data.csv) | column -t -s','

Tabulate the output of a loop

for f in /var/log/*.log; do echo "$(wc -l < "$f") $(du -h "$f" | cut -f1) $f"; done | column -t

Format paste output into a clean table

paste <(cut -d: -f1 /etc/passwd) <(cut -d: -f3 /etc/passwd) <(cut -d: -f7 /etc/passwd) | column -t

Side-by-side lists — combine paste with column

paste <(seq 5) <(printf '%s\n' a b c d e) | column -t

Wrap long lines into columns — fill terminal width

ls /usr/bin | column

Control number of columns with -c (terminal width in characters)

ls /usr/bin | column -c 80