Using Regular Expressions in Visual Studio

Visual Studio uses .NET Framework regular expressions to find and replace text. In Visual Studio 2010 and earlier versions, Visual Studio used custom regular expression syntax in the Find and Replace windows. This topic explains how to convert some of the more commonly-used custom regular expression symbols to the .NET versions.

Tip

In Windows operating systems, most lines end in “\r\n” (a carriage return followed by a new line). These characters are not visible, but are present in the editor and are passed to the .NET Regular Expression service.

For more information, see .NET Framework Regular Expressions.

Tip

For information about regular expressions that are used in replacement patterns, see Substitutions in Regular Expressions. To use a numbered capture group, the syntax is $1 to specify the numbered group and (x) to specify the group in question:. For example, the grouped regular expression (\d)([a-z]) finds four matches in the following string: 1a 2b 3c 4d. The replacement string z$1 converts that string to z1 z2 z3 z4> The equivalent syntax in Visual Studio 2010 is {:z}([a-z]) for the grouped regular expression and z\1 for the replacement string.

Comparing.NET Framework Regular Expressions

You can convert regular expressions from the version used in previous versions of Visual Studio to .NET Framework regular expressions by using the following table.

Purpose

New

Old

New Example

Match any single character (except a line break)

.

.

a.o matches "aro" in "around" and "abo" in "about" but not "acro" in "across".

Match zero or more occurrences of the preceding expression (match as many characters as possible)

*

*

a*r matches "r" in "rack", "ar" in "ark", and "aar" in "aardvark"

Match any character zero or more times (Wildcard *)

.*

.*

c.*e matches “cke” in “racket”, “comme” in “comment”, and “code” in “code”

Match one or more occurrences of the preceding expression (match as many characters as possible)

+

+

e.+e matches "eede" in "feeder" but not "ee".

Match any character one or more times (Wildcard ?)

.+

.+

e.+e matches "eede" in "feeder" but not "ee".

Match zero or more occurrences of the preceding expression (match as few characters as possible)

*?

@

e.*?e matches "ee" in "feeder" but not "eede".

Match one or more occurrences of the preceding expression (match as few characters as possible)

+?

#

e.+?e matches "ente" and "erprise" in "enterprise", but not the whole word "enterprise".

Anchor the match string to the beginning of a line or string

^

^

^car matches the word "car" only when it appears at the beginning of a line.

Anchor the match string to the end of a line

\r?$

$

End\r?$ matches "end" only when it appears at the end of a line.

Match any single character in a set

[abc]

[abc]

b[abc] matches "ba", "bb", and "bc".

Match any character in a range of characters

[a-f]

[x-y]

be[n-t] matches "bet" in "between", "ben" in "beneath", and "bes" in "beside", but not "below".

Capture and implicitly number the expression contained within parenthesis

()

()

([a-z])X\1 matches "aXa"and "bXb", but not "aXb". ". “\1” refers to the first expression group “[a-z]”.

Invalidate a match

(?!abc)

~(abc)

real (?!ity) matches "real" in "realty" and "really" but not in "reality." It also finds the second "real" (but not the first "real") in "realityreal".

Match any character that is not in a given set of characters

[^abc]

[^abc]

be[^n-t] matches "bef" in "before", "beh" in "behind", and "bel" in "below", but not "beneath".

Match either the expression before or the one after the symbol.

|

|

(sponge|mud) bath matches "sponge bath" and "mud bath."

Escape the character following the backslash

\

\

\^ matches the character ^.

Specify the number of occurrences of the preceding character or group

{x}, where x is the number of occurrences

\x, where x is the number of occurrences

x(ab){2}x matches "xababx", and x(ab){2,3}x matches "xababx" and "xabababx" but not "xababababx".

Match text in a Unicode character class, where “X” is the Unicode number. For more information about Unicode character classes, see

Unicode Standard 5.2 Character Properties.

\p{X}

:X

\p{Lu} matches "T" and "D" in "Thomas Doe".

Match a word boundary

\b (Outside a character class \b specifies a word boundary, and inside a character class specifies a backspace).

< and > specify the beginning and end of a word

\bin matches "in" in "inside" but not "pinto".

Match a line break (ie a carriage return followed by a new line).

\r?\n

\n

End\r?\nBegin matches "End" and "Begin" only when "End" is the last string in a line and "Begin" is the first string in the next line.

Match any alphanumeric character

\w

:a

a\wd matches "add" and "a1d" but not "a d".

Match any whitespace character.

(?([^\r\n])\s)

:b

Public\sInterface matches the phrase "Public Interface".

Match any numeric character

\d

:d

\d matches and "3" in "3456", "2" in 23", and "1" in "1".

Match a Unicode character

\uXXXX where XXXX specifies the Unicode character value.

\uXXXX where XXXX specifies the Unicode character value

\u0065 matches the character "e".

Match an identifier

\b(_\w+|[\w-[0-9_]]\w*)\b

:i

Matches "type1" but not &type1" or "#define".

Match a string inside quotes

((\".+?\")|('.+?'))

:q

Matches any string inside single or double quotes.

Match a hexadecimal number

\b0[xX]([0-9a-fA-F])\b

:h

Matches "0xc67f" but not "0xc67fc67f".

Match integers and decimals

\b[0-9]*\.*[0-9]+\b

:n

Matches "1.333".

See Also

Other Resources

Finding and Replacing Text