Minimal — zero or more

@

Matches zero or more occurrences of the preceding expression, matching as few characters as possible.

Minimal — one or more

#

Matches one or more occurrences of the preceding expression, matching as few characters as possible.

Repeat n times

^n

Matches n occurrences of the preceding expression. For example, [0-9]^4 matches any 4-digit sequence.

Grouping

()

Groups a subexpression.

nth tagged text

\n

In a Find or Replace expression, indicates the text matched by the nth tagged expression, where n is a number from 1 to 9.

In a Replace expression, \0 inserts the entire matched text.

Right-justified field

\(w,n)

In a Replace expression, right-justifies the nth tagged expression in a field at least w characters wide.

Left-justified field

\(-w,n)

In a Replace expression, left-justifies the nth tagged expression in a field at least w characters wide.

Prevent match

~(X)

Prevents a match when X appears at this point in the expression. For example, real~(ity) matches the "real" in "realty" and "really," but not the "real" in "reality."

Alphanumeric character

:a

Matches the expression ([a-zA-Z0-9]).

Alphabetic character

:c

Matches the expression ([a-zA-Z]).

Decimal digit

:d

Matches the expression ([0-9]).

Hexadecimal digit

:h

Matches the expression ([0-9a-fA-F]+).

Rational number

:n

Matches the expression (([0-9]+.[0-9]*)|([0-9]*.[0-9]+)|([0-9]+)).

Alphabetic string

:w

Matches the expression ([a-zA-Z]+).

Escape

\e

Unicode U+001B.

Bell

\g

Unicode U+0007.

Backspace

\h

Unicode U+0008.

Tab

\t

Matches a tab character, Unicode U+0009.

Unicode character

\x#### or \u####

Matches a character given by Unicode value where #### is hexadecimal digits. You can specify a character outside the Basic Multilingual Plane (that is, a surrogate) with the ISO 10646 code point or with two Unicode code points giving the values of the surrogate pair.