OK, I just saw David Cassell wrote an excellent answer, but I had already
started composing this before leaving last night, so I'll send this off
anyway. Hope it gives a little additional insight.
Andrew Staples wrote:
> if ($zone !~ m%^(?:((?:\w[-A-Za-z0-9]*\.)+\w\w+)|((?:(?:[12]
> [0-9][0-9]|[1-9][0-9]?)\.){2,4})[Ii][Nn]-[Aa][Dd][Dd][Rr]\.[Aa]
> [Rr][Pp][Aa]\.?)$%x){
Let's split this up into multiple lines, to make it more legible:
if($zone !~ m% # match -- using m%foo%x instead of /foo/x
^ # beginning of line
(?: # grouping (non-capturing) parentheses
( # capture to $1
(?: # grouping parentheses
\w # one "alphanumunder"
[-A-Za-z0-9]* # zero or more of hyphen, letter, or number
\. # followed by dot
)+ # one or more of above
\w\w+ # followed by two or more alphanumunders
) # end of capturing
| # alternative
( # capture to $2
(?: # grouping parens
(?: # more grouping parens
[12][0-9][0-9] # number from 100 to 299
| # or
[1-9][0-9]? # number 1-9, or 10-99
) # end of grouping
\. # one dot
){2,4} # above two to four times
) # end of capturing
[Ii][Nn]-[Aa][Dd][Dd][Rr]\.[Aa][Rr][Pp][Aa]
# 'in-addr.arpa', case insensitively
\.? # optional dot
) # end of grouping
$ # end of line
%x) { # note /x switch: "free-form" regexes allowed
# i.e. whitespace and comments are ignored
# some processing
}
> m = treat the string as multiple lines (not sure why this
> should happen)
No, that would be /m , as in an m after the match ends. Here you have a
match where you can choose your own delimiter. /foo/ = m!foo! = m{foo} =
m%foo% etc.
> %=mod, doesn't make sense to me
see above.
> ^ = beginning of string
yup.
> ?: = pattern match up to : *or* right operator
No: (?: .... ) are parentheses that are used for grouping but not for
capturing.
> \w[-A-Za-z0-9] = look for character that isn't alpha (is that
> what the leading '-' is?) up to the first '.'
No. \w is a "word" character (letters, numbers, underscore) and a dash at
the beginning of a character class is just interpreted as a literal dash.
'Not a letter or digit' would be [^A-Za-z0-9] with a caret at the beginning.
> I pretty much get lost after that. I think its verifying
> that the the first 3 numbers must be 1 or 2, the second and
> third three must be 0-9.
No, it's two to four repetitions of (a number from 1 to 299, inclusive).
> I'm not sure what the '|' symbol is for either.
It separates alternatives: /foo|bar/ matches either foo or bar.
Hope this helps. You may also want to review perlop and perlre on the syntax
of m// and regular expressions, respectively.
Cheers,
Philip
---
You are currently subscribed to perl-win32-users as: [archive@jab.org]
To unsubscribe, forward this message to
[EMAIL PROTECTED]
For non-automated Mailing List support, send email to
[EMAIL PROTECTED]
