Regex cheat sheet

Character classes

Pattern	Matches
.	Any single character except newline (use the s flag if you also want newlines).
\d	A digit. By default in ICU/NSRegularExpression this is 0-9 only; turn on Unicode for full digits.
\w	A "word" character: letters, digits, underscore.
\s	Any whitespace: space, tab, newline.
[abc]	One of a, b, or c.
[^abc]	Anything that is not a, b, or c.
[a-z]	Any lowercase ASCII letter. Use [a-zA-Z] for both cases or just add the i flag.

Quantifiers

Pattern	Meaning
?	0 or 1 (optional).
*	0 or more.
+	1 or more.
{3}	Exactly 3.
{2,5}	Between 2 and 5.
{2,}	2 or more.
+?, *?	Lazy versions. Match as little as possible instead of as much as possible.

Lazy versus greedy is the source of more wrong matches than anything else. If you write ".*" against the string "foo" and "bar", you get the whole thing because * is greedy. ".*?" gives you just "foo".

Anchors and boundaries

Pattern	Meaning
^	Start of the string (or line, with the m flag).
$	End of the string (or line, with m).
\b	Word boundary. Useful for whole-word matches: \bcat\b matches "cat" but not "category".
\B	Anywhere that is not a word boundary.

Groups

Parentheses do two things at once: they group, and they capture.

Pattern	Meaning
(abc)	Group and capture, accessible as $1 or \1.
(?:abc)	Group only, no capture. Use this when you just need | or quantifiers and do not want extra capture slots.
(?<name>abc)	Named capture. Reference with \k<name> or ${name}.
a|b	Alternation: a or b.

Lookahead and lookbehind

These check for context without consuming characters. Read once, you will never forget them.

Pattern	Meaning
(?=foo)	Followed by "foo". a(?=b) matches the "a" in "ab" but not in "ac".
(?!foo)	Not followed by "foo".
(?<=foo)	Preceded by "foo".
(?<!foo)	Not preceded by "foo".

Flags

Flag	Effect
i	Case insensitive.
m	Multiline: ^ and $ match start/end of each line.
s	Dotall: . also matches newlines.
u	Unicode: full Unicode awareness for \d, \w, etc.
x	Extended: ignore whitespace and allow comments inside the pattern.

Patterns you will probably need

Email-ish (catches almost anything that looks like an email, not RFC-strict):

[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}

URL:

https?://[^\s]+

UK / US phone, very loose:

\+?\d[\d\s().-]{7,}\d

ISO date YYYY-MM-DD:

\b\d{4}-\d{2}-\d{2}\b

IPv4 (loose):

\b(?:\d{1,3}\.){3}\d{1,3}\b

Hex color:

#?[0-9a-fA-F]{6}\b

Trim multiple blank lines into one (replace mode, replacement = \n\n):

(\r?\n){3,}

The escape rules

Most regex flavours treat these as special characters and require a backslash if you want the literal version: . ^ $ * + ? ( ) [ ] { } | \. Inside a character class ([ ]), most of these are already literal, but ] \ and - (in the middle of a class) still need attention.

When you copy a regex into a JSON string or a JavaScript source file, every backslash has to be doubled. \d in your pattern becomes "\\d" in JSON. This is the second-most common reason a working regex stops working when pasted somewhere.