Regular Expression Engine Comparison Chart

regular_expression_engine_comparison.md

Regular Expression Engine Comparison Chart

Many different applications claim to support regular expressions. But what does that even mean?

Well there are lots of different regular expression engines, and they all have different feature sets and different time-space efficiencies.

The information here is just copied from: http://regular-expressions.mobi/refflavors.html

But for some reason, it's not accessible unless you have a mobile phone user agent.

Go to the main site for lots of regular expression information and their commercial product called RegexBuddy.

Regular Expression Flavors (Engines)

• JGsoft: This flavor is used by the Just Great Software products, including PowerGREP and EditPad Pro.
• .NET: This flavor is used by programming languages based on the Microsoft .NET framework versions 1.x, 2.0 or 3.x. It is generally also the regex flavor used by applications developed in these programming languages.
• Java: The regex flavor of the java.util.regex package, available in the Java 4 (JDK 1.4.x) and later. A few features were added in Java 5 (JDK 1.5.x) and Java 6 (JDK 1.6.x). It is generally also the regex flavor used by applications developed in Java.
• Perl: The regex flavor used in the Perl programming language, versions 5.6 and 5.8. Versions prior to 5.6 do not support Unicode.
• PCRE: The open source PCRE library. The feature set described here is available in PCRE 5.x and 6.x. PCRE is the regex engine used by the TPerlRegEx Delphi component and the RegularExrpessions and RegularExpressionsCore units in Delphi XE and C++Builder XE.
• ECMA (JavaScript): The regular expression syntax defined in the 3rd edition of the ECMA-262 standard, which defines the scripting language commonly known as JavaScript.
• Python: The regex flavor supported by Python's built-in re module.
• Ruby: The regex flavor built into the Ruby programming language.
• Tcl ARE: The regex flavor developed by Henry Spencer for the regexp command in Tcl 8.2 and 8.4, dubbed Advanced Regular Expressions.
• POSIX BRE: Basic Regular Expressions as defined in the IEEE POSIX standard 1003.2.
• POSIX ERE: Extended Regular Expressions as defined in the IEEE POSIX standard 1003.2.
• GNU BRE: GNU Basic Regular Expressions, which are POSIX BRE with GNU extensions, used in the GNU implementations of classic UNIX tools.
• GNU ERE: GNU Extended Regular Expressions, which are POSIX ERE with GNU extensions, used in the GNU implementations of classic UNIX tools.
• XML: The regular expression flavor defined in the XML Schema standard.
• XPath: The regular expression flavor defined in the XQuery 1.0 and XPath 2.0 Functions and Operators standard.

Applications Implementing a Regular Expression Flavor

• AceText: Version 2 and later use the JGsoft engine. Version 1 did not support regular expressions at all.
• awk: The awk UNIX tool and programming language uses POSIX ERE.
• C#: As a .NET programming language, C# can use the System.Text.RegularExpressions classes, listed as ".NET" below.
• Delphi for .NET: As a .NET programming language, the .NET version of Delphi can use the System.Text.RegularExpressions classes, listed as ".NET" below.
• Delphi for Win32: Delphi for Win32 does not have built-in regular expression support. Many free PCRE wrappers are available.
• EditPad Pro: Version 6 and later use the JGsoft engine. Earlier versions used PCRE, without Unicode support.
• egrep: The traditional UNIX egrep command uses the "POSIX ERE" flavor, though not all implementations fully adhere to the standard. Linux usually ships with the GNU implementation, which use "GNU ERE".
• grep: The traditional UNIX grep command uses the "POSIX BRE" flavor, though not all implementations fully adhere to the standard. Linux usually ships with the GNU implementation, which use "GNU BRE".
• Emacs: The GNU implementation of this classic UNIX text editor uses the "GNU ERE" flavor, except that POSIX classes, collations and equivalences are not supported.
• Java: The regex flavor of the java.util.regex package is listed as "Java" in the table below.
• JavaScript: JavaScript's regex flavor is listed as "ECMA" in the table below.
• MySQL: MySQL uses POSIX Extended Regular Expressions, listed as "POSIX ERE" in the table below.
• Oracle: Oracle Database 10g implements POSIX Extended Regular Expressions, listed as "POSIX ERE" in the table below. Oracle supports backreferences \1 through \9, though these are not part of the POSIX ERE standard.
• Perl: Perl's regex flavor is listed as "Perl" in the table below.
• PHP: PHP's ereg functions implement the "POSIX ERE" flavor, while the preg functions implement the "PCRE" flavor.
• PostgreSQL: PostgreSQL 7.4 and later uses Henry Spencer's "Advanced Regular Expressions" flavor, listed as "Tcl ARE" in the table below. Earlier versions used POSIX Extended Regular Expressions, listed as POSIX ERE.
• PowerGREP: Version 3 and later use the JGsoft engine. Earlier versions used PCRE, without Unicode support.
• PowerShell: PowerShell's built-in -match and -replace operators use the .NET regex flavor. PowerShell can also use the System.Text.RegularExpressions classes directly.
• Python: Python's regex flavor is listed as "Python" in the table below.
• R: The regular expression functions in the R language for statistical programming use either the POSIX ERE flavor (default), the PCRE flavor (perl = true) or the POSIX BRE flavor (perl = false, extended = false).
• REALbasic: REALbasic's RegEx class is a wrapper around PCRE.
• RegexBuddy: Version 3 and later use a special version of the JGsoft engine that emulates all the regular expression flavors in this comparison. Version 2 supported the JGsoft regex flavor only. Version 1 used PCRE, without Unicode support.
• Ruby: Ruby's regex flavor is listed as "Ruby" in the table below.
• sed: The sed UNIX tool uses POSIX BRE. Linux usually ships with the GNU implementation, which use "GNU BRE".
• Tcl: Tcl's Advanced Regular Expression flavor, the default flavor in Tcl 8.2 and later, is listed as "Tcl ARE" in the table below. Tcl's Extended Regular Expression and Basic Regular Expression flavors are listed as "POSIX ERE" and "POSIX BRE" in the table below.
• VBScript: VBScript's RegExp object uses the same regex flavor as JavaScript, which is listed as "ECMA" in the table below.
• Visual Basic 6: Visual Basic 6 does not have built-in support for regular expressions, but can easily use the "Microsoft VBScript Regular Expressions 5.5" COM object, which implements the "ECMA" flavor listed below.
• Visual Basic.NET: As a .NET programming language, VB.NET can use the System.Text.RegularExpressions classes, listed as ".NET" below.
• wxWidgets: The wxRegEx class supports 3 flavors. wxRE_ADVANCED is the "Tcl ARE" flavor, wxRE_EXTENDED is "POSIX ERE" and wxRE_BASIC is "POSIX BRE".
• XML Schema: The XML Schema regular expression flavor is listed as "XML" in the table below.
• XPath: The regex flavor used by XPath functions is listed as "XPath" in the table below.
• XQuery: The regex flavor used by XQuery functions is listed as "XPath" in the table below.

Feature Comparison

Characters
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
Backslash escapes one metacharacter	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
\Q...\E escapes a string of metacharacters	YES	no	Java 6	YES	YES	no	no	no	no	no	no	no	no	no	no
\x00 through \xFF (ASCII character)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	no	no	no	no
\n (LF), \r (CR) and \t (tab)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	no	no	YES	YES
\f (form feed) and \v (vtab)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	no	no	no	no
\a (bell)	YES	YES	YES	YES	YES	no	YES	YES	YES	no	no	no	no	no	no
\e (escape)	YES	YES	YES	YES	YES	no	no	YES	YES	no	no	no	no	no	no
\b (backspace) and \B (backslash)	no	no	no	no	no	no	no	no	YES	no	no	no	no	no	no
\cA through \cZ (control character)	YES	YES	YES	YES	YES	YES	no	no	YES	no	no	no	no	no	no
\ca through \cz (control character)	YES	YES	no	YES	YES	YES	no	no	YES	no	no	no	no	no	no
Character Classes or Character Sets [abc]
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
[abc] character class	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
[^abc] negated character class	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
[a-z] character class range	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
Hyphen in [\d-z] is a literal	YES	YES	YES	YES	YES	no	no	no	no	no	no	no	no	no	no
Hyphen in [a-\d] is a literal	YES	no	no	no	YES	no	no	no	no	no	no	no	no	no	no
Backslash escapes one character class metacharacter	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	no	no	YES	YES
\Q...\E escapes a string of character class metacharacters	YES	no	Java 6	YES	YES	no	no	no	no	no	no	no	no	no	no
\d shorthand for digits	YES	YES	ascii	YES	ascii	ascii	option	ascii	YES	no	no	no	no	YES	YES
\w shorthand for word characters	YES	YES	ascii	YES	ascii	ascii	option	ascii	YES	no	no	YES	YES	YES	YES
\s shorthand for whitespace	YES	YES	ascii	YES	ascii	YES	option	ascii	YES	no	no	YES	YES	ascii	ascii
\D, \W and \S shorthand negated character classes	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	YES	YES	YES	YES
[\b] backspace	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	no	no	no	no
Dot
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
. (dot; any character except line break)	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
Anchors
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
^ (start of string/line)	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	YES
$ (end of string/line)	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	YES
\A (start of string)	YES	YES	YES	YES	YES	no	YES	YES	YES	no	no	no	no	no	no
\Z (end of string, before final line break)	YES	YES	YES	YES	YES	no	no	YES	YES	no	no	no	no	no	no
\z (end of string)	YES	YES	YES	YES	YES	no	\Z	YES	no	no	no	no	no	no	no
\` (start of string)	no	no	no	no	no	no	no	no	no	no	no	YES	YES	no	no
\' (end of string)	no	no	no	no	no	no	no	no	no	no	no	YES	YES	no	no
Word Boundaries
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
\b (at the beginning or end of a word)	YES	YES	YES	YES	ascii	ascii	option	ascii	no	no	no	YES	YES	no	no
\B (NOT at the beginning or end of a word)	YES	YES	YES	YES	ascii	ascii	option	ascii	no	no	no	YES	YES	no	no
\y (at the beginning or end of a word)	YES	no	no	no	no	no	no	no	YES	no	no	no	no	no	no
\Y (NOT at the beginning or end of a word)	YES	no	no	no	no	no	no	no	YES	no	no	no	no	no	no
\m (at the beginning of a word)	YES	no	no	no	no	no	no	no	YES	no	no	no	no	no	no
\M (at the end of a word)	YES	no	no	no	no	no	no	no	YES	no	no	no	no	no	no
\< (at the beginning of a word)	no	no	no	no	no	no	no	no	no	no	no	YES	YES	no	no
\> (at the end of a word)	no	no	no	no	no	no	no	no	no	no	no	YES	YES	no	no
Alternation
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
| (alternation)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	YES	\|	YES	YES	YES
Quantifiers
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
? (0 or 1)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	YES	\?	YES	YES	YES
* (0 or more)	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
+ (1 or more)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	YES	\+	YES	YES	YES
{n} (exactly n)	YES	YES	YES	YES	YES	YES	YES	YES	YES	\{n\}	YES	\{n\}	YES	YES	YES
{n,m} (between n and m)	YES	YES	YES	YES	YES	YES	YES	YES	YES	\{n,m\}	YES	\{n,m\}	YES	YES	YES
{n,} (n or more)	YES	YES	YES	YES	YES	YES	YES	YES	YES	\{n,\}	YES	\{n,\}	YES	YES	YES
? after any of the above quantifiers to make it "lazy"	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	no	no	no	YES
Grouping and Backreferences
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
(regex) (numbered capturing group)	YES	YES	YES	YES	YES	YES	YES	YES	YES	\( \)	YES	\( \)	YES	YES	YES
(?:regex) (non-capturing group)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	no	no	no	no
\1 through \9 (backreferences)	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	YES	YES	no	YES
\10 through \99 (backreferences)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	n/a	no	no	n/a	YES
Forward references \1 through \9	YES	YES	YES	YES	YES	no	no	YES	no	no	n/a	no	no	n/a	no
Nested references \1 through \9	YES	YES	YES	YES	YES	YES	no	YES	no	no	n/a	no	no	n/a	no
Backreferences non-existent groups are an error	YES	YES	YES	YES	YES	no	YES	no	YES	YES	n/a	YES	YES	n/a	YES
Backreferences to failed groups also fail	YES	YES	YES	YES	YES	no	YES	YES	YES	YES	n/a	YES	YES	n/a	YES
Modifiers
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
(?i) (case insensitive)	YES	YES	YES	YES	YES	/i only	YES	YES	YES	no	no	no	no	no	flag
(?s) (dot matches newlines)	YES	YES	YES	YES	YES	no	YES	(?m)	no	no	no	no	no	no	flag
(?m) (^ and $ match at line breaks)	YES	YES	YES	YES	YES	/m only	YES	always on	no	no	no	no	no	no	flag
(?x) (free-spacing mode)	YES	YES	YES	YES	YES	no	YES	YES	YES	no	no	no	no	no	flag
(?n) (explicit capture)	YES	YES	no	no	no	no	no	no	no	no	no	no	no	no	no
(?-ismxn) (turn off mode modifiers)	YES	YES	YES	YES	YES	no	no	YES	no	no	no	no	no	no	no
(?ismxn:group) (mode modifiers local to group)	YES	YES	YES	YES	YES	no	no	YES	no	no	no	no	no	no	no
Atomic Grouping and Possessive Quantifiers
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
(?>regex) (atomic group)	YES	YES	YES	YES	YES	no	no	YES	no	no	no	no	no	no	no
?+, *+, ++ and {m,n}+ (possessive quantifiers)	YES	no	YES	no	YES	no	no	no	no	no	no	no	no	no	no
Lookaround
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
(?=regex) (positive lookahead)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	no	no	no	no
(?!regex) (negative lookahead)	YES	YES	YES	YES	YES	YES	YES	YES	YES	no	no	no	no	no	no
(?<=text) (positive lookbehind)	full regex	full regex	finite length	fixed length	fixed + alternation	no	fixed length	no	no	no	no	no	no	no	no
(?<!text) (negative lookbehind)	full regex	full regex	finite length	fixed length	fixed + alternation	no	fixed length	no	no	no	no	no	no	no	no
Continuing from The Previous Match
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
\G (start of match attempt)	YES	YES	YES	YES	YES	no	no	YES	no	no	no	no	no	no	no
Conditionals
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
(?(?=regex)then|else) (using any lookaround)	YES	YES	no	YES	YES	no	no	no	no	no	no	no	no	no	no
(?(regex)then|else)	no	YES	no	no	no	no	no	no	no	no	no	no	no	no	no
(?(1)then|else)	YES	YES	no	YES	YES	no	YES	no	no	no	no	no	no	no	no
(?(group)then|else)	YES	YES	no	no	YES	no	YES	no	no	no	no	no	no	no	no
Comments
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
(?#comment)	YES	YES	no	YES	YES	no	YES	YES	YES	no	no	no	no	no	no
Free-Spacing Syntax
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
Free-spacing syntax supported	YES	YES	YES	YES	YES	no	YES	YES	YES	no	no	no	no	no	YES
Character class is a single token	YES	YES	no	YES	YES	n/a	YES	YES	YES	n/a	n/a	n/a	n/a	n/a	YES
# starts a comment	YES	YES	YES	YES	YES	n/a	YES	YES	YES	n/a	n/a	n/a	n/a	n/a	no
Unicode Characters
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
\X (Unicode grapheme)	YES	no	no	YES	option	no	no	no	no	no	no	no	no	no	no
\u0000 through \uFFFF (Unicode character)	YES	YES	YES	no	no	YES	u"string"	no	YES	no	no	no	no	no	no
\x{0} through \x{FFFF} (Unicode character)	YES	no	no	YES	option	no	no	no	no	no	no	no	no	no	no
Unicode Properties, Scripts and Blocks
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
\pL through \pC (Unicode properties)	YES	no	YES	YES	option	no	no	no	no	no	no	no	no	no	no
\p{L} through \p{C} (Unicode properties)	YES	YES	YES	YES	option	no	no	no	no	no	no	no	no	YES	YES
\p{Lu} through \p{Cn} (Unicode property)	YES	YES	YES	YES	option	no	no	no	no	no	no	no	no	YES	YES
\p{L&} and \p{Letter&} (equivalent of [\p{Lu}\p{Ll}\p{Lt}] Unicode properties)	YES	no	no	YES	option	no	no	no	no	no	no	no	no	no	no
\p{IsL} through \p{IsC} (Unicode properties)	YES	no	YES	YES	no	no	no	no	no	no	no	no	no	no	no
\p{IsLu} through \p{IsCn} (Unicode property)	YES	no	YES	YES	no	no	no	no	no	no	no	no	no	no	no
\p{Letter} through \p{Other} (Unicode properties)	YES	no	no	YES	no	no	no	no	no	no	no	no	no	no	no
\p{Lowercase_Letter} through \p{Not_Assigned} (Unicode property)	YES	no	no	YES	no	no	no	no	no	no	no	no	no	no	no
\p{IsLetter} through \p{IsOther} (Unicode properties)	YES	no	no	YES	no	no	no	no	no	no	no	no	no	no	no
\p{IsLowercase_Letter} through \p{IsNot_Assigned} (Unicode property)	YES	no	no	YES	no	no	no	no	no	no	no	no	no	no	no
\p{Arabic} through \p{Yi} (Unicode script)	YES	no	no	YES	option	no	no	no	no	no	no	no	no	no	no
\p{IsArabic} through \p{IsYi} (Unicode script)	YES	no	no	YES	no	no	no	no	no	no	no	no	no	no	no
\p{BasicLatin} through \p{Specials} (Unicode block)	YES	no	no	YES	no	no	no	no	no	no	no	no	no	no	no
\p{InBasicLatin} through \p{InSpecials} (Unicode block)	YES	no	YES	YES	no	no	no	no	no	no	no	no	no	no	no
\p{IsBasicLatin} through \p{IsSpecials} (Unicode block)	YES	YES	no	YES	no	no	no	no	no	no	no	no	no	YES	YES
Part between {} in all of the above is case insensitive	YES	no	no	YES	no	no	no	no	no	no	no	no	no	no	no
Spaces, hyphens and underscores allowed in all long names listed above (e.g. BasicLatin can be written as Basic-Latin or Basic_Latin or Basic Latin)	YES	no	Java 5	YES	no	no	no	no	no	no	no	no	no	no	no
\P (negated variants of all \p as listed above)	YES	YES	YES	YES	option	no	no	no	no	no	no	no	no	YES	YES
\p{^...} (negated variants of all \p{...} as listed above)	YES	no	no	YES	option	no	no	no	no	no	no	no	no	no	no
Named Capture and Backreferences
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
(?<name>regex) (.NET-style named capturing group)	YES	YES	no	no	no	no	no	no	no	no	no	no	no	no	no
(?'name'regex) (.NET-style named capturing group)	YES	YES	no	no	no	no	no	no	no	no	no	no	no	no	no
\k<name> (.NET-style named backreference)	YES	YES	no	no	no	no	no	no	no	no	no	no	no	no	no
\k'name' (.NET-style named backreference)	YES	YES	no	no	no	no	no	no	no	no	no	no	no	no	no
(?P<name>regex) (Python-style named capturing group	YES	no	no	no	YES	no	YES	no	no	no	no	no	no	no	no
(?P=name) (Python-style named backreference)	YES	no	no	no	YES	no	YES	no	no	no	no	no	no	no	no
multiple capturing groups can have the same name	YES	YES	n/a	n/a	no	n/a	no	n/a	n/a	n/a	n/a	n/a	n/a	n/a	n/a
XML Character Classes
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
\i, \I, \c and \C shorthand XML name character classes	no	no	no	no	no	no	no	no	no	no	no	no	no	YES	YES
[abc-[abc]] character class subtraction	YES	2.0	no	no	no	no	no	no	no	no	no	no	no	YES	YES
POSIX Bracket Expressions
Feature	JGsoft	.NET	Java	Perl	PCRE	ECMA	Python	Ruby	Tcl ARE	POSIX BRE	POSIX ERE	GNU BRE	GNU ERE	XML	XPath
[:alpha:] POSIX character class	YES	no	no	YES	ascii	no	no	YES	YES	YES	YES	YES	YES	no	no
\p{Alpha} POSIX character class	YES	no	ascii	no	no	no	no	no	no	no	no	no	no	no	no
\p{IsAlpha} POSIX character class	YES	no	no	YES	no	no	no	no	no	no	no	no	no	no	no
[.span-ll.] POSIX collation sequence	no	no	no	no	no	no	no	no	YES	YES	YES	YES	YES	no	no
[=x=] POSIX character equivalence	no	no	no	no	no	no	no	no	YES	YES	YES	YES	YES	no	no

I think java support, named-capturing groups 🤔

I have basic knowledge of regex. It seems that there is no single compliant implementation as of 2022. It seems to me that maybe they are supersets of the standards. Maybe those in charge of the standardization are not catching up fast enough with the industry or they refuse to include features when there is some minimal agreement among the developers of these software products. What does not look good is the fact that none is identical to another. That means that developers and users must be always aware of this when building and using, respectively, applications that use regex and that also that inter-operate or interact with other applications.
The problem is greater for users as implementations can be (and usually are) black boxes for them.
Even if two people use the same product, let's say the same web browser, but version "a" of the browser uses a certain "flavor" of the regex engine and, version "b" of the browser uses another flavor or, simply introduced a change or whatever. They are not going to get the same result if the expression requires a feature that causes the engines to yield different results. T E R R I B L E !!!
But, if the application does not interact with anything that requires a matching engine, well, there is a benefit, a great one. Certain tasks can be done very quickly with such a compact notation.
Thanks to all the contributors. This page is UNIQUE and extremely HELPFUL. The people in charge of the standardization of regex should pay you all.

This comparison table is being very useful. Thank you very much!

Is this document for ECMA a daydream?

Wow, thank you so much for documenting this. I do not see the comparison table on the regular-expressions.info website linked, even on mobile view. Is this now really the only freely-available reference that compares & outlines the features of each regex flavor?

I'm willing to help lead an effort to get this documented and hosted on a domain name. This information is too important to be lost!

VBScript: VBScript's RegExp object uses the same regex flavor as JavaScript, which is listed as "ECMA" in the table below.

I'm not sure that's true anymore... There are many things VBScript does not allow which ECMA does I think...?

In fact, the regular expression flavor used in the version 5.5 VBScript object is the same one used by JavaScript and JScript. The regex flavor is part of the ECMA-262 standard for JavaScript. Therefore, everything said about JavaScript’s regular expression flavor in this book also applies to VBScript. JavaScript and VBScript implement Perl-style regular expressions. However, they lack quite a number of advanced features available in Perl and other modern regular expression flavors:
• No \A or \Z anchors to match the start or end of the string. Use a caret or dollar instead.
• Lookbehind is not supported at all. Lookahead is fully supported.
• No atomic grouping or possessive quantifiers
• No Unicode support, except for matching single characters with \uFFFF
• No named capturing groups. Use numbered capturing groups instead.
• No mode modifiers to set matching options within the regular expression.
• No conditionals.
• No regular expression comments.
Describe your regular expression with VBScript apostrophe comments instead, outside the regular expression string.
Version 1.0 of the RegExp object even lacks basic features like lazy quantifiers. This is the main reason this book does not discuss VBScript RegExp 1.0. All versions of Internet Explorer prior to 5.5 include version 1.0 of the RegExp object. There are no other versions than 1.0 and 5.5.

Src: Regular Expressions - The Complete Tutorial - Jan Goyvaerts

Hi, what about re2?

Perhaps add "what characters does \w match for each language."

What about Raku, the language formerly known as Perl6? Larry Wall and the rest of the crew re-wrote Perl's regex implementation to be more intuitive/powerful in Perl6. See:

https://docs.raku.org/language/regexes
https://docs.raku.org/
https://raku.org

Rakudo, the principal Raku interpreter/compiler, was released in 2015.
See: https://www.rakudo.org

Thanks for making this. The Java part seems outdated. Java has named capturing groups since Java 7. There's also \R that matches all line ends (LF, CR, and the different combinations in use) - I don't know if that exists in other regex dialects though.

Since I could not find any page listing the changes, I asked AI Assistant, I hope the information is valid and can help you updating this gist. This is what I got:

Here are some notable changes and additions to the Java Regex Implementation by various versions of Java:

• Java 5.0 (2004): Named capturing groups feature is added. You can name the capturing groups in the form (?<name>X). Named-capturing group is still available via the group index.

  Example: Pattern.compile("(?<area>\\d+)-(?<prefix>\\d+)-(?<line>\\d+)")

• Java 7 (2011): The UNICODE_CHARACTER_CLASS flag is introduced. This makes \b, \s, \w, \B, \S, \W, \d, \D match unicode characters, not only ASCII.

  Example: Pattern.compile("\\w", Pattern.UNICODE_CHARACTER_CLASS)

• Java 8 (2014): A final new feature added to the regular expressions was the word boundary \b that now works on the Unicode.

• Java 9 (2017): Introduction of \R as universal line end matcher. You can use \R to match any linebreak sequence.

  Example: Pattern.compile("\\R")

  This will match these line endings: \u000D\u000A|[\u000A\u000B\u000C\u000D\u001C\u001D\u001E\u0085\u2028\u2029]

This information is obtained from the different Java API documentations for these versions. Let me know if you need further information about how to use them etc.