Regex Tester

Which flavour of regex is this?

JavaScript's, because it runs in your browser. That matters — regex dialects genuinely differ, and a pattern that works here may not work identically in PowerShell (.NET), grep (POSIX), Python or PCRE.

The differences that bite most often: lookbehind (?<=...) works in modern JavaScript and .NET but not in POSIX grep; named groups are (?<name>...) in JavaScript and .NET but (?P<name>...) in Python; and \d is Unicode-aware in some engines and ASCII-only in others.

The patterns worth knowing

• \b — word boundary. The fix for a pattern that matches inside longer words.
• (?:...) — group without capturing. Use it whenever you only need the grouping, not the value.
• +? and *? — lazy quantifiers. The answer when a greedy .* swallows the whole line.
• ^ and $ with the m flag — anchor to each line rather than the whole string. Essential for log parsing.

A word of caution about validating things with regex

Regex is excellent at extracting and terrible at validating anything with real grammar. The IP pattern in the example above matches 999.999.999.999 quite happily — it checks shape, not validity. Email is worse: the fully correct RFC 5322 pattern is thousands of characters long and still does not tell you the address exists.

For those, match loosely to extract, then validate properly in code. Our subnet calculator does real IP validation rather than a shape check, which is why it rejects 255.255.0.255 as a mask.

Catastrophic backtracking

Nested quantifiers over the same characters — the classic being (a+)+$ — can take exponential time on input that nearly matches. That is a real denial-of-service vector if the pattern runs on user input server-side. If a pattern here appears to hang, that is what you are looking at, and it is worth knowing before it ships.