Regex Lookahead and Lookbehind, Explained With Real Examples

Lookahead and lookbehind (together called "lookaround") let a regex check what comes before or after a position in the text without including that part in the actual match — the engine checks the condition, then doesn't move. That's the one idea behind all four variants: positive, negative, ahead, and behind.

Positive lookahead: (?=...)

q(?=u) matches a q, but only when it's followed by a u — and the u itself isn't part of the match. A practical example: \d+(?=px) matches the number in "20px" without capturing the unit, useful when you want the value but not the label attached to it.

Negative lookahead: (?!...)

X(?!Y) means "match X, but only if it's not followed by Y." A common use: matching a word that isn't immediately followed by a specific suffix, like flagging a filename that isn't already suffixed with .test before appending it yourself.

Positive lookbehind: (?<=...)

(?<=\$)\d+(?:\.\d{2})? matches the 19.99 inside $19.99, leaving the dollar sign itself out of the match entirely — the lookbehind confirms a $ precedes the number without consuming it. This is the pattern behind extracting a price or an ID that always follows a fixed label.

Negative lookbehind: (?<!...)

(?<!a)b matches a "b" that isn't preceded by an "a" — it matches the b in "rob" (preceded by o) but skips the b in "cab" (preceded by a). Negative lookbehind is the tool for "match this, but only when NOT immediately after a specific thing."

Why this matters in practice

Lookaround turns validation you'd otherwise need multiple passes or a much bigger pattern for into a single expression — currency parsing, filtering matches by context, and pulling a value out from between fixed labels all lean on it. The one tradeoff: lookbehind support varies more across regex engines and older JavaScript versions than lookahead does, so test the exact pattern in the environment you'll actually run it in.