Regex Syntax Reference

Any character except newline.

a.c → 'abc', 'a_c'

Digit — equivalent to [0-9].

\d+ → '123'

Non-digit.

\D+ → 'abc'

Word character — [a-zA-Z0-9_].

\w+ → 'hello_42'

Non-word character.

\W → ' ', '!'

Whitespace — space, tab, newline.

\s+ between words

Non-whitespace.

\S+ → 'hello'

Any character in the set.

[aeiou] → 'a', 'e'

Any character NOT in the set.

[^aeiou] → consonants

Any character in the range.

[a-z]+ → 'hello'

0 or more of the preceding element.

a* → '', 'a', 'aaa'

1 or more of the preceding element.

a+ → 'a', 'aaa'

0 or 1 of the preceding element — makes it optional.

colou?r → 'color', 'colour'

Exactly n repetitions.

\d{4} → '2024'

At least n repetitions.

\d{3,} → '123', '1234'

Between n and m repetitions.

\d{2,4} → '12', '1234'

Lazy (non-greedy) quantifiers — match as few as possible.

<.+?> matches inner tag only

Start of string (or line in multiline mode).

^Hello → 'Hello World'

End of string (or line in multiline mode).

end$ → 'the end'

Word boundary — between \w and \W.

\bcat\b → 'cat' but not 'catch'

Non-word boundary.

\Bcat → 'catch'

Start of string (Python/not JS).

End of string (Python/not JS).

Capturing group — captures the match for back-references.

(ab)+ → 'ababab'

Non-capturing group — groups without capturing.

(?:https?://)\S+

Named capturing group.

(?<year>\d{4})-(?<month>\d{2})

Back-reference to capturing group 1, 2 etc.

(.)\1 → 'aa', 'bb'

Alternation — matches a or b.

cat|dog → 'cat', 'dog'

Positive lookahead — asserts abc follows.

\d+(?= dollars)

Negative lookahead — asserts abc does NOT follow.

\d+(?! dollars)

Positive lookbehind — asserts abc precedes.

(?<=\$)\d+

Negative lookbehind — asserts abc does NOT precede.

(?<!\$)\d+

Case-insensitive matching.

/hello/i → 'Hello', 'HELLO'

Global — find all matches, not just the first.

/a/g → all 'a' in string

Multiline — ^ and $ match start/end of each line.

/^hello/m

Dotall — . matches newline too.

/.+/s

Extended (verbose) — ignore whitespace, allow comments. (Python/Ruby)

/ # match digits
\d+/x

Unicode mode — full Unicode matching (JS/Python).

/\p{L}+/u