Regular Expression

Anchors:
^            Start of string    
\A           Start of string
$            End of string
\Z           End of string 
\b           Word boundary
\B           Not word boundary
\<           Start of word
\>           End of word

Character Classes:
\c           Control character
\s           White space
\S           Not white space
\d           Digit
\D           Not digit
\w           Word
\W           Not word
\x           Hexadecimal digit
\O           Octal digit

POSIX:
[:upper:]    Upper case letters
[:lower:]    Lower case letters
[:alpha:]    All letters
[:alnum:]    Digits and letters
[:digit:]    Digits
[:xdigit:]   Hexadecimal digits
[:punct:]    Punctuation
[:blank:]    Space and tab
[:space:]    Blank characters
[:cntrl:]    Control characters
[:graph:]    Printed characters
[:print:]    Printed characters and spaces
[:word:]     Digits, letters and underscore

Assertions:
?=          Lookahead assertion
?!          Negative lookahead
?<=         Lookbehind assertion
?!= or ?<!  Negative lookbehind
?>          Once-only Subexpression
?()         Condition [if then]
?()|        Condition [if then else]
?#          Comment

Quantifiers:
*           0 or more
+           1 or more
!           Do not match the next character or regular expression.
{3}         Exactly 3
{3,}        3 or more
{3,5}       3, 4 or 5

Quantifier Modifiers:
"x" below represents a quantifier
x?        Ungreedy version of "x"

Ecape Character:
\         Escape Character

Metacharacters (must be escaped):
^     $      (        )       <
.     *      +        ?       [
{     \      |        >

Special Characters:
\n        New line
\r        Carriage return
\t        Tab
\v        Vertical tab
\f        Form feed
\xxx      Octal character xxx
\xhh      Hex character hh

Groups and Ranges:
.         Any character except new line (\n)
(a|b)     a or b
(...)     Group
(?:...)   Passive Group
[abc]     Range (a or b or c)
[^abc]    Not a or b or c
[a-m]     Letter between a and m
[A-M]     Upper case letter between A and M
[0-5]     Digit between 0 and 5
\n        nth group/subpattern
Note: Ranges are inclusive.

Pattern Modifiers:
g         Global match
i         Case-insensitive
m         Multiple lines
s         Treat string as single line
x         Allow comments and white space in pattern
e         Evaluate replacement
U         Ungreedy pattern

String Replacement (Backreferences):
$n        nth non-passive group
$2        "xyz" in /^(abc(xyz))$/
$1        "xyz" in /^(?:abc)(xyz)$/
$`        Before matched string
$'        After matched string
$+        Last matched string
$&        Entire matched string

Sample Patterns:
Pattern                                     Will Match
([A-Za-z0-9-]+)                             Letters, numbers and hyphens
(\d{1,2}\/\d{1,2}\/\d{4})                   Date (e.g. 21/3/2006)
([^\s]+(?=\.(jpg|gif|png))\.\2)             jpg, gif or png image
(^[1-9]{1}$|^[1-4]{1}[0-9]{1}$|^50$)        Any number from 1 to 50 inclusive
(#?([A-Fa-f0-9]){3}(([A-Fa-f0-9]){3})?)     Valid hexadecimal colour code
((?=.*\d)(?=.*[a-z])(?=.*[A-Z]).{8,15})     String with at least one upper case letter, one lower
                                            case letter, and one digit (useful for passwords).
(\w+@[a-zA-Z_]+?\.[a-zA-Z]{2,6})            Email addresses
(\<(/?[^\>]+)\>)                            HTML Tags

Note: These patterns are intended for reference purposes and have not been extensively tested. 
Please use with caution and test thoroughly before use.