Regular Expressions for Mere Mortals

A regular expression (regex) is a tiny language for describing text patterns: "an email-ish thing," "a date," "every error line in this log." One line of regex replaces dozens of lines of manual string-chopping code — which is why every programmer eventually learns them, usually the hard way.

This guide is the easy way. No CS degree required. Ideas first, then patterns you can try immediately.

What this guide covers

What regex is — and the famous warning about two problems
First patterns — literal matches and the dot
Character classes — [abc], \d, \w, \s
Quantifiers — *, +, ?, {n,m}, greedy vs. lazy
Anchors — ^, $, \b
Groups — capturing, backreferences, non-capturing groups
Alternation — this|that
Escaping — matching literal special characters
Flags — case-insensitivity, multiline, dotall
Real-world recipes — dates, logs, CSV-ish data, validation vs. extraction
Flavors — why the same pattern behaves differently in Python, JavaScript, and grep
Troubleshooting — catastrophic backtracking and debugging

What it doesn't cover

Parser theory, writing your own regex engine, or the full PCRE spec. This guide teaches the 20% of regex that handles 95% of real tasks — and tells you when to stop and write real code instead.

How to use this guide

Every example is runnable. Python examples use the re module; shell examples use grep -E. Check: blocks show verified output. Try it: marks patterns worth typing into a regex tester (regex101.com explains each token as you type — use it while reading this guide).

A note on verification

Every pattern in this guide was executed against Python 3.12's re module (and grep -E where shown) before shipping — 75 assertions, all green. Where behavior differs between regex flavors, it's called out explicitly instead of smoothed over.

Chapters

Written to the nine-point writing standard shared by every guide in this series.