regexp to get path from URL

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: regexp to get path from URL by chromatic (Archbishop) on Sep 29, 2000 at 21:50 UTC
The URI module has a path() method... could come in handy. If you know the path is sane, you might do something like this: `(my $path = $ENV{HTTP_REFERRER}) =~ s!//(.+?)/[^/]+\??!$1!;` I would rather have a module do it, though I've also split on the forward slashes... Update:* In the interests of historical accuracy, I'll simply humour Ovid by providing a more workable regex: `(my $path = $ENV{HTTP_REFERER}) =~ s!\.\w+/(.+?)/[^/]+\?*?!$1!;` This is your brain on work.	[reply] [d/l] [select]
(Ovid) Re: regexp to get path from URL by Ovid (Cardinal) on Sep 29, 2000 at 22:10 UTC
chromatic's regex appears to be picking up the domain rather than the path. Did I miss something? Heh. chromatic misspelled 'HTTP_REFERER' by spelling 'referrer' correctly :) The beauty of misspelled standards... I'd use the following regex, so long as you remember that the referrer can be easily spoofed and this isn't being used for security: `$ENV{HTTP_REFERER} =~ m!^[^:]+://[^/]+/([^?]+)! or die "No path info!" +; my $path = $1;` [download] The final `[^?]+` strips off the query string. Switch it to `.*` if you want the query string. The `or die` on the regex kicks off if we don't have a path. Obviously, if the referrer is something like "http://www.yahoo.com/", you're going to have a problem. However, I assumed that your referrer would probably have a path appended. You can change that to `or some_error_subroutine()` if you like. Alternately, if no path info is acceptable, make sure that $path is set to "" on a regex failure. Otherwise, if anything was already in $1, then that would be assigned to $path and obviously, this is undesirable (thanks to chromatic for pointing that out to me). Cheers, Ovid Join the Perlmonks Setiathome Group or just go the the link and check out our stats.	[reply] [d/l]