Regular Expressions cheat sheet
Basic matching
Each symbol matches a single character:
.anything1
\d digit in 0123456789
\D non-digit
\w “word” (letters and digits and _)
\W non-word
␣ space
\t tab
\r return
\n new line2
\s whitespace (␣, \t,\r,\n)
\S non-whitespace
Character classes
Character classes [...] match any of the charac-
ters in the class. Ex: [aeiou] matches vowels.
Use ˆto specify the complement set: [ˆaeiou]
matches non-vowels (including non-letters!).
Use -to specify a range of letters:
[a-e] matches abcde and [0-9a-f] matches
‘0123456789abcdef‘.
Boundaries
Boundary characters are helpful in “anchor-
ing” your pattern to some edge, but do not
select any characters themselves.
\b word boundaries (defined as any edge
between a \w and \W)
\B non-word-boundaries
ˆthe beginning of the line
$the end of the line
Ex: \bcat \b finds a match in “the cat in the hat”
but not in “locate”.
Disjunction
(X|Y) X or Y
Ex: \b(cat|dog)s\b matches cats and dogs.
“Quantifiers”
X* 0 or more repetitions of X
X+ 1 or more repetitions of X
X? 0 or 1 instances of X
X{m}exactly minstances of X
X{m,} at least minstances of X
X{m,n}between mand n(inclusive) in-
stances of X
By default, quantifiers just apply to the one
character. Use (...)to specify explicit quanti-
fier “scope.”
Ex: ab+ matches ab,abb,abbb,abbbb...
(ab)+ matches ab,abab,ababab...
Quantifiers are by default greedy in regex.
Good regex engines support adding ?to a
quantifier to make it lazy.
Ex: greedy: ^.*b aabaaba
lazy: ^.*?b aabaaba
Special characters
The characters {}[]()^$.|*+?\ (and -inside
[...]) have special meaning in regex, so they
must be “escaped” with \to match them.
Ex: \. matches the period .and \\ matches the
backslash \.
Backreferences
Count your open parentheses (from the left,
starting with 1. Whatever is matched byparen-
thesis number ncan be referenced later by \n.
Ex: \b(\w+)␣\1\b matches two identical words
with a space in between
Backreferences are useful for find/replaces:
Ex: Finding \b (\w +)er\b and replacing with
more \1 will map “the taller man” 7→ “the
more tall man” and “I am shorter” 7→ “I am
more short”.
Advanced
Read about “non-capturing parentheses” and
“look-ahead” and “look-behind” online. Also,
visualize your regexes as finite-state machines
at http://www.regexper.com/.
1...except line breaks, depending on your engine.
2Depending on where you got your file, line breaks may be \r,\n, or \r\n. Also, in some regex engines
(e.g. TextWrangler), \r and \n match the same things.
Hadas Kotek (courtesy of Michael Y. Erlewine)