Regex Cheatsheet

The syntax of regular expressions on one page, each item with an example you can open in the Regex Tester, and ready-made patterns written out for six languages.

Open Regex Tester Type a pattern and some text to see what it matches.

Characters and classes

.	Any character except a line break (with `s`, line breaks too).	c.t (open in Regex Tester)
\d	A digit. In JavaScript only `0-9`; in Python 3 any Unicode digit.	\d+ (open in Regex Tester)
\w	A word character: letters, digits, underscore (ASCII in JavaScript).	\w+ (open in Regex Tester)
\s	Whitespace: space, tab, line breaks and more.	\s+ (open in Regex Tester)
\D \W \S	The opposite of `\d`, `\w`, `\s`.	\D+ (open in Regex Tester)
[abc]	One of the listed characters.	gr[ae]y (open in Regex Tester)
[^abc]	Any character except the listed ones.	[^aeiou\s]+ (open in Regex Tester)
[a-z]	A range. Combine them: `[A-Za-z0-9_]`.	[A-F0-9]+ (open in Regex Tester)
\b	A word boundary: between `\w` and something else.	\bcat\b (open in Regex Tester)

Anchors

^	Start of the string (with `m`, of each line).	^\w+ (open in Regex Tester)
$	End of the string (with `m`, of each line).	\w+$ (open in Regex Tester)
\A \z	Start and end of the whole string, whatever the flags, in Java, PHP, Go and .NET. Python writes the end as `\Z` (`\z` too from 3.14). JavaScript has neither; use `^` and `$` without `m`.

Quantifiers

*	Zero or more.	ab*c (open in Regex Tester)
+	One or more.	ab+c (open in Regex Tester)
?	Zero or one: makes the item optional.	colou?r (open in Regex Tester)
{3}	Exactly 3.	\d{3} (open in Regex Tester)
{2,}	2 or more.	\d{2,} (open in Regex Tester)
{2,4}	From 2 to 4.	\d{2,4} (open in Regex Tester)
*? +? ??	Lazy: as few as possible. `<.+?>` stops at the first `>`.	<.+?> (open in Regex Tester)

Groups and lookaround

(abc)	A capturing group: matched text is kept as `$1`, `\1`, `group(1)`.	(\d{4})-(\d{2}) (open in Regex Tester)
(?:abc)	A group that does not capture: for alternation and quantifiers.	(?:ab)+ (open in Regex Tester)
(?<year>\d{4})	A named group. Python's `re` writes it `(?P<year>...)`.	(?<year>\d{4}) (open in Regex Tester)
a\|b	Either side.	cat\|dog (open in Regex Tester)
\1	A backreference: the text group 1 matched. Not in Go.	(\w)\1 (open in Regex Tester)
(?=...)	Followed by, without consuming it. Not in Go.	\d+(?=px) (open in Regex Tester)
(?!...)	Not followed by. Not in Go.	\d+(?!px\|\d) (open in Regex Tester)
(?<=...)	Preceded by. Not in Go.	(?<=\$)\d+ (open in Regex Tester)
(?<!...)	Not preceded by. Not in Go.	(?<![\$\d])\d+ (open in Regex Tester)

Flags in JavaScript

g	Global: find every match, not just the first.
i	Ignore case.
m	Multiline: `^` and `$` match at line breaks.
s	Dot-all: `.` matches line breaks too.
u	Unicode: code points instead of UTF-16 units, and `\p{...}` classes.
v	Like `u`, plus set operations in classes. Use one of `u` or `v`.
y	Sticky: match only at `lastIndex`.
d	Indices: report where each group starts and ends.

Characters to escape

Outside a class, put a backslash before any of . * + ? ^ $ ( ) [ ] { } | \ to match it literally, and before / inside a JavaScript literal. Inside [...], only ] \ ^ - need it, and only where they would otherwise mean something. With the v flag, a class also needs a backslash before ( ) [ ] { } / - |, and doubled punctuation such as && or !! is reserved.

Flavors differ

Go (RE2) has no lookaround and no backreferences. In exchange every match runs in linear time.
Python's re names groups (?P<name>...); JavaScript, Java and .NET write (?<name>...).
\d is 0-9 in JavaScript, but any Unicode digit in Python 3 and .NET unless you ask for ASCII.
Python's re.match anchors only at the start; re.fullmatch and Java's matches() test the whole string.

Ready-made patterns

16 guides