The catch: it may be followed by many options -- a colon, ^Note_1 if it's followed by a port number; a questionmark for several possible uses; and perhaps others that I'm blanking on just now. But regardless, the entity from the last slash, through a period to the next punctuation should be what you're looking for.

And to broaden the hint a bit further, the regex documentation and tutorials here will show you precisely the way to obtain what you're looking for.

Update: ^Note_1 See correction (+ + by quester immediately below. Aargh.

... a colon, if it's followed by a port number...

Minor nit: The colon and port number is just after the hostname in a URL, not the page name. For example, consider the port 8080 in

http://www.example.com:8080/pagename.html

The question mark following the page name in a URL starts a list of parameters being passed from the browser to the script running in the server. The parameter values can be more or less anything; by convention spaces will have been replaced by plus signs, but otherwise almost anything goes, including colons. For example,

http://www.example.com/filename.pl?credentials=myuserid:zomg_dont_send_passwords_in_the_clear

my ($pagename) = $url =~ m{([^/]+)$};
[download]

... but has a (greedy) failure mode:

C:\>perl -e "my $url = 'http://www.perlmonks.com/index.pl?node_id=9674
+84'; my ($pagename) = $url =~ m{([^/]+)$}; print $pagename;"
index.pl?node_id=967484
C:\>
[download]

Thanks everyone for the quick replies. This regular expression is just what I needed to see what I was doing wrong.

Of course test the URL that you are trying to get by using your normal Web browser. If it can't "get" it, Perl can't either.