This SLB method creates a policy that match the url with the simplified regular expression <regex> defined.

<regex> is a string of the form:

[^]string1[*string2[*string3...[*stringN]..]][$]

The so-called "meta" characters in the regular expression have the
following meanings:

^ - Match at the beginning of the URL
* - Wildcard - any sequence of 0 or more characters will match
$ - Match at the end of the URL

Example regular expressions would be:

^I=*you=*me=*.html$ (matches I=Jamie/you=Arhtur/index.html, but not
youandI=me/you=vinny/index.html or
I=Jamie/you=Arthur/index.htmlamentation)

^I=*you=*me=*.html (matches I=Jamie/you=Arthur/index.htmlamentation)

I=*you=*me=*.html (matches youandI=me/you=vinny/index.html)

*.html$ (matches any URL ending in .html)

Invalid regular expressions include:

^^hi$ (more than one caret (^) is not allowed)
^hi$$ (more than one dollar ($) is not allowed)
h^i (caret must be at the start)
$hi (dollar must be at the end)
^*hi (consecutive meta characters are not allowed)
h**i (consecutive meta characters are not allowed)
* (must contain at least one non-meta character)
^ (Ibid)
$ (Ibid)

If the regular expression matches the URL, then the policy matches.

<group_name> should correspond to a group with an method other than
"pc" or "pu". <precedence> can be any value between 0 and 65535 inclusive.
Up to 50 regex policies can be configured per virtual.