Regex cheatsheet
Many programs use regular expression to find & replace text. However, they tend to come with their own different flavor.
You can probably expect most modern software and programming languages to be using some variation of the Perl flavor, "PCRE"; however command-line tools (grep, less, ...) will often use the POSIX flavor (sometimes with an extended variant, e.g. egrep or sed -r). Vim also comes with its own syntax (a superset of what Vi accepts).
This cheatsheet lists the respective syntax of each flavor, and the software that uses it.
If you spot errors or missing data, or just want to make this prettier/more accurate, don't hesitate to open an issue or a pull request.
Syntax
| What | Perl/PCRE | Python's re | POSIX (BRE) | POSIX extended (ERE) | Vim |
|---|---|---|---|---|---|
| Basics | |||||
| Custom character class | [...] | [...] | [...] | [...] | [...] |
| Negated custom character class | [^...] | [^...] | [^...] | [^...] | [^...] |
| \ special in class? | yes | yes | no, ] escaped if comes first | no, ] escaped if comes first | yes |
| Ranges | [a-z], - escaped if first or last | [a-z], - escaped if first or last | [a-z], - escaped if first or last | [a-z], - escaped if first or last | [a-z], - escaped if first or last |
| Alternation | | | | | \| | | | \| \& (low precedence) |
| Escaped character | \033 \x1B \x{1234} \N{name} \N{U+263D} | \x12 | \%d123 \%x2A \%u1234 \%U1234ABCD | ||
| Character classes | |||||
| Any character (except newline) | . | . | . | . | . |
| Any character (including newline) | \_. | ||||
Match a "word" character (alphanumeric plus _) | \w [[:word:]] | \w | \w | \w | \w |
| Case | [[:upper:]] / [[:lower:]] | [[:upper:]] / [[:lower:]] | [[:upper:]] / [[:lower:]] | \u [[:upper:]] / \l [[:lower:]] | |
| Match a non-"word" character | \W | \W | \W | ||
| Match a whitespace character (except newline) | \s [[:space:]] | \s [[:space:]] | \s [[:space:]] | ||
| Whitespace including newline | \s [[:space:]] | \s | \_s | ||
| Match a non-whitespace character | \S | \S | [^[:space:]] | [^[:space:]] | \S [^[:space:]] |
| Match a digit character | \d [[:digit:]] | \d | [[:digit:]] | [[:digit:]] | \d [[:digit:]] |
| Match a non-digit character | \D | \D | [^[:digit:]] | [^[:digit:]] | \D [^[:digit:]] |
| Any hexadecimal digit | [[:xdigit:]] | [[:xdigit:]] | [[:xdigit:]] | \x [[:xdigit:]] | |
| Any octal digit | \o | ||||
| Any graphical character excluding "word" characters | [[:punct:]] | [[:punct:]] | [[:punct:]] | [[:punct:]] | |
| Any alphabetical character | [[:alpha:]] | [[:alpha:]] | [[:alpha:]] | \a [[:alpha:]] | |
| Non-alphabetical character | [^[:alpha:]] | [^[:alpha:]] | \A [^[:alpha:]] | ||
| Any alphanumerical character | [[:alnum:]] | [[:alnum:]] | [[:alnum:]] | [[:alnum:]] | |
| ASCII | [[:ascii:]] | ||||
| Character equivalents (e = é = è) (as per locale) | [[=e=]] | [[=e=]] | [[=e=]] | ||
| Zero-width assertions | |||||
| Word boundary | \b | \b | \b / \< (start) / \> (end) | \b / \< (start) / \> (end) | \< (start) / \> (end) |
| Anywhere but word boundary | \B | \B | \B | \B | |
| Beginning of line/string | ^ / \A | ^ / \A | ^ | ^ | ^ (beginning of pattern ) \_^ |
| End of line/string | $ / \Z | $ / \Z | $ | $ | $ (end of pattern) \_$ |
| Captures and groups | |||||
| Capturing group | (...) (?<name>...) | (...) (?P<name>...) | \(...\) | (...) | \(...\) |
| Non-capturing group | (?:...) | (?:...) | \%(...\) | ||
| Backreference to a specific group. | \1 \g1 \g{-1} | \1 | \1 | \1 non-official | \1 |
| Named backreference | \g{name} \k<name> | (?P=name) | |||
| Look-around | |||||
| Positive look-ahead | (?=...) | (?=...) | \(...\)\@= | ||
| Negative look-ahead | (?!...) | (?!...) | \(...\)\@! | ||
| Positive look-behind | (?<=...) | (?<=...) | \(...\)\@<= | ||
| Negative look-behind | (?<!...) | (?<!...) | \(...\)\@<! | ||
| Multiplicity | |||||
| 0 or 1 | ? | ? | \? | ? | \? |
| 0 or more | * | * | * | * | * |
| 1 or more | + | + | \+ | + | \+ |
| Specific number | {n} {n,m} {n,} | {n} {n,m} {n,} | \{n\} \{n,m\} \{n,\} | {n} {n,m} {n,} | \{n} \{n,m} \{n,} |
| 0 or 1, non-greedy | ?? | ?? | |||
| 0 or more, non-greedy | *? | *? | \{-} | ||
| 1 or more, non-greedy | +? | +? | |||
| Specific number, non-greedy | {n,m}? {n,}? | {n,m}? {n,}? | \{-n,m} \{-n,} | ||
| 0 or 1, don't give back on backtrack | ?+ | ||||
| 0 or more, don't give back on backtrack | *+ | ||||
| 1 or more, don't give back on backtrack | ++ | ||||
| Specific number, don't give back on backtrack | {n,m}+ {n,}+ | ||||
| Other | |||||
| Independent non-backtracking pattern | (?>...) | \(...\)\@> | |||
| Make case-sensitive/insensitive | (?i) / (?-i) | (?i) / (?-i) | \c / \C | ||
Programs
| What | Syntax | Comments/gotchas |
|---|---|---|
| Programming languages | ||
| Perl | PCRE | PCRE is actually a separate implementation from Perl's, with slight differences |
Python's re standard lib | Python's own syntax (Perl-inspired) | |
| Ruby | Ruby's own syntax (Perl-inspired) | |
| Java's java.util.regex | Almost PCRE | |
| Boost.Regex | PCRE | |
| Text editors | ||
| Eclipse | PCRE | |
| Emacs | ? | |
| Netbeans | PCRE | |
| Notepad++ | PCRE (Boost.Regex) | |
| PyCharm | PCRE | Perl-inspired |
| Sublime Text | ? | |
| UltraEdit | PCRE | |
| Vim | Vim | |
| Command-line tools | ||
| awk | ERE | might depend on the implementation |
| grep | BRE, egrep for ERE, grep -P for PCRE (optional) | |
| less | ERE | usually; man page says "regular expression library supplied by your system" |
| screen | plain text | |
| sed | BRE, -E switches to ERE | |