Roman Sokolyuk wrote:
Thank you very much for your time.
This fixed the problem.
I have only one question. How does the browser determine which parts of the
URL to strip?
It has very precise rules for that, which are probably to be found in
the HTTP RFC (http://www.faqs.org/rfcs/rfc2616.html), or maybe the URL
RFC (http://www.ietf.org/rfc/rfc2396.txt).
Rather daunting documents, but anyone fiddling with WWW development
should at least read them once (repeat 3 times) (*)
I hope nobody bites me for the following, but here is a short summary
off-the-cuff :
- split the full URL from which the current page was obtained (say
http://somehost.somedomain.com/somedir/someapp?some=query
into
protocol: http
(separator): ://
hostname: somehost.somedomain.com
path: /somedir/someapp
query: some=query
- keep only the path for now
- strip the last path component (say, work backward stripping
characters from the end until the first /)
- add any relative reference to that path
- re-assemble a URL with the protocol, host, and new path
- go get it
(*) and anyone fiddling with Tomcat should add the Servlet Specification.
(If I hadn't added this, Chuck would have)
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
