Negative Lookahead — (?!...)

Assert that something does NOT follow

(?!...) is the inverse: "at this position, the pattern inside does NOT match." Same zero-width behavior — no characters consumed. The match succeeds only if the lookahead pattern would FAIL.

Common patterns

Match a word UNLESS it's followed by a specific suffix. cat(?!s) matches 'cat' but not 'cats' — the negative lookahead asserts 's' does not come next.

Find tokens that aren't followed by a stop word. Useful for parsing where you want "any word EXCEPT when followed by reserved words."

Avoid false positives in URL detection. http(?!s://) matches 'http' but not 'https://' — useful when you want to upgrade plain HTTP references but skip already-secure ones.

The mental flip

Most beginners struggle with negative lookahead because it requires inverting the question. Instead of "what should match?" you ask "what should the next thing NOT be?" Practice helps; the cleanest patterns often use a positive description in the main match plus a negative lookahead to exclude one specific case.

Code

Negative lookahead patterns·python

import re

# 'cat' but not 'cats'
re.findall(r'\bcat(?!s)\b', 'cats and a cat scatter')
# ['cat']  — only the standalone 'cat'

# Plain 'http' but not 'https://'
re.findall(r'http(?!s://)', 'http://a.com https://b.com httpd')
# ['http', 'http']  — first http and the 'http' in httpd

# Number not followed by 'rd' (so not '3rd')
re.findall(r'\b\d+(?!rd\b)', '1st 2nd 3rd 4th 7 11')
# ['1', '2', '4', '7', '11']  — '3' excluded because 'rd' follows

# Words that aren't followed by 'ing'
re.findall(r'\b\w+(?!ing)\b', 'run running jump jumping read')
# Note: this is tricky because \w+ is greedy — adjust pattern as needed

Negative Lookahead — (?!...)

Assert that something does NOT follow

Common patterns

The mental flip

Code

External links

Exercise

Progress

Comments 0