1 Introduction

Regular expression (regexp or regex) is a sequence of characters that define a search pattern, comes from formal language theory. and supported by IEEE Posix with sets BRE(Basic Regular Expression) and ERE(Extended Regular Expression).

2 Posix metacharacters

2.1 Basic regular expression

MetaCh_BRE	Description
^S	matches the string start with S (of any line)
S$	matches the string end with S (of any line)
[S]	bracket expression. Matches a single character in the scope of S ([S1-S2], [S1-], [-S2], ...)
[^S]	complement of [S]
.	matches any single character, except '\n' line-break
S*	matches S zero to gigantic times
S{m, n}	matches S at least m and not more than n times
(exp)	sub-expression
\n	n $\subseteq$ [1, 9], refer to which sub-expression

'S', 'S1', 'S2' refer string or single character

2.2 Extended regular expression

MetaCh_ERE	Description
(exp)	same as (exp)
S{m, n}	same as S{m, n}
S?	matches S zero or one time
+	matches S one to gigantic times
S1\|S2	optional S1 or S2

'S', 'S1', 'S2' refer string or single character

Tips : Because of its expressive power and (relative) ease of reading, many other utilities and programming languages have adopted syntax similar to Perl's — for example, Java, JavaScript, Julia, Python, Ruby, Qt, Microsoft's .NET Framework, and XML Schema, PHP, awk, sed, etc.

3 Character classes

The character class is a quick way to express the expressions set.

POSIX	Perl	Vim	ASCII	Description
[:alnum:]	-	-	[A-Za-z0-9]	Alphanumeric characters
[:alpha:]	-	\a	[A-Za-z]	Alphabetic characters
[:blank:]	-	\s	[ \t]	Space and tab
[:cntrl:]	-	-	[\x00-\x1F\x7F]	Control characters
[:digit:]	\d	\d	[0-9]	Digits
[:lower:]	-	\l	[a-z]	Lowercase letters
[:upper:]	-	\u	[A-Z]	Uppercase letters
[:print:]	-	\p	[\x20-\x7E]	Visible characters and the space character
[:space:]	\s	_s	[ \t\r\n\v\f]	Whitespace characters
[:xdigit:]	-	\x	[A-Fa-f0-9]	Hexadecimal digits

comparison of different standards about character class

3 Lazy and possessive matching

4 Unicode

5 Wildcard Vs RegExp

Wildcards are placeholders while regular expressions are searching pattern, and only '.' in regexp is a wildcard.

MetaCh	Description
*	placeholder for any characters, zero - gigantic
?	placeholder for any single character

wildcards meta-ch list

Tips : some implementations also contain meta-ch similar to regexp, such as [S], [!S] (which equal to [^S]).

RegExp