sort — sort lines of text

What it does

sort reads lines of text from a file or stdin and writes them back in sorted order. By default it sorts lexically (byte-wise, like a dictionary), but with -n it compares numbers, with -k it sorts by a specific field, and with -r it reverses the order. Because it works on streams, it is the natural partner of every other text tool in the pipeline.

Why it matters

sort is what turns raw output into something readable. Want the biggest files first, the most common log message, or an alphabetical index? sort is the first step. Teamed with uniq (to collapse duplicates) and head or tail (to trim to the extremes), it becomes a tiny data-analysis engine that runs entirely on the command line with no spreadsheet in sight.

Example

$ sort /etc/passwd | head -3

daemon:x:1:1:daemon:/usr/sbin:/usr/sbin/nologin
bin:x:2:2:bin:/bin:/usr/sbin/nologin
nobody:x:65534:65534:nobody:/nonexistent:/usr/sbin/nologin

$ du -h --max-depth=1 | sort -h -r | head -5

12G	./node_modules
1.2G	./.git
340M	./dist
 22M	./src
4.0K	./.env

Common flags

-n

sort numerically instead of lexically (so 10 comes after 2)

-r

reverse the sort order

-k N

sort by the Nth whitespace-separated field

-t CH

set the field separator character (used with -k), e.g. -t: for /etc/passwd

-h

sort by human-readable sizes like 2K, 300M, 12G

-u

output only unique lines (like sort then uniq)

-f

fold case — ignore case differences while sorting