sort: Bring Order to Chaos

You have a file with data in random order. You want it sorted. That's literally what sort does:

sort names.txt

Alphabetical order. Done.

Alphabetical vs Numerical

Here's where people get tripped up. By default, sort orders alphabetically:

That's "correct" alphabetically (1 comes before 2, and 10 starts with 1). But if you want numerical order:

sort -n numbers.txt

Now you get 1, 2, 3, 10.

Reverse Order

sort -r file.txt

Z to A, or highest to lowest with -rn.

Sorting by Column

Real data has multiple columns. Sort by the second one:

sort -k2 data.txt

Sort by the third column numerically:

sort -k3 -n scores.txt

Custom Delimiters

If your columns aren't space-separated:

sort -t',' -k2 data.csv

The -t',' sets comma as the delimiter.

Remove Duplicates While Sorting

sort -u file.txt

Like sort | uniq but in one command.

The Classic Pipeline

This pattern appears everywhere:

cat data.txt | sort | uniq -c | sort -rn | head

Read → sort → count duplicates → sort by count → show top results.

It answers "what are the most common items?"

Real Examples

Sort a leaderboard by score (highest first):

sort -k2 -rn scores.txt

Sort /etc/passwd by user ID:

sort -t':' -k3 -n /etc/passwd

Find most frequent log entries:

cat access.log | sort | uniq -c | sort -rn | head -10

Quick Reference

| What you want | Command | |---------------|---------| | Alphabetical | sort file | | Numerical | sort -n file | | Reverse | sort -r file | | By column | sort -k2 file | | Numeric by column | sort -k2 -n file | | With delimiter | sort -t',' -k2 file | | Unique only | sort -u file |

Practice

sort is fundamental in CTF challenges for frequency analysis and organizing data for the next step in your pipeline.

sort is the setup for everything else. Once your data is in order, patterns emerge.