Regular Expressions

Refer to the Perl documentation for regular expressions for more information.

Escape sequences for common classes of characters

\d

Matches any digit

\w

Matches a-z, A-Z, 0-9, _

\s

Matches spaces, tabs

\b

Matches a word boundary (not necessarily a space: foo.bar is two words)

\D

Opposite of \d

\W

Opposite of \w

\S

Opposite of \s

\B

Opposite of \b

Special symbols for matching

\

Escape the special meaning if the following character

.

Match any character. \. matches a period.

$

Mathces the end of line character

^

Matches the start of a line. ^$ will match an empty line.

|

Alternation: match one or the other. one|two will match either the word one or two.

[]

Character class. Matches any one of the characters in the brackets: [two] will match one of the letters t, w, or o. A hyphen means a range of character when it is inside the brackets: [0-9]matches any digit. Note: if the hypen is first, then it means a hyphen, not a range indicator: [-09] matches hypen, 0, or 9.

[^]

Negate the character class. Matches any one of the characters that is not in the brackets: [^two] will match any one letter, except the letters t, w, or o. Note: ^ does not match the start of a line when it is inside [].

( )

Remember the subpattern that matches the pattern inside the parentheses. The subpatterns are stored in special varaibles: $1, $2, $3, etc. These can be referenced until the next regular expressions changes them. ([hj]ello) will match either hello or jello and store the matched word in $1.

(?: )

Same as ( ), but don't set the $ variables. Used to group patterns.

Special symbols for repetition

*

Match 0 or more occurences of the preceding pattern

+

Match 1or more occurences of the preceding pattern

?

Match 0 or 1 occurences of the preceding pattern

{n}

Match exactly n occurences of the preceding pattern

{m,n}

Match from m to n occurences occurences of the preceding pattern.

{,n}

Match from 0 to n occurences of the preceding pattern. (? = {,1})

{m,}

Match m or more occurences of the preceding pattern. (* = {0,}; + = {1,})

A few examples

fred|barney|wilma|betty

Matches any one of the words fred, barney, wilma, betty

Miami, (?:FL|OH)

Matches Miami, FL or Miami, OH. Does not set $1. The pattern could also be Miami, (FL|OH), in which case $1 would be either FL or OH. The grouping is necessary, otherwise the pattern would match Miami, FL or OH.

[cm][[ao][dp]

Matches cad, cap, cod, cop, mad, map, mod, mop.

\d{3}-\d{2}-\d{4}

Matches a social security number

^\s*(\w+)\s*$

Matches an entire line. Looks for a word in the line, and stores the matched word in $1.