From: James Britt <jamesUNDERBARb@...>
Date: 2004-09-14T02:33:57+09:00
Subject: Re: [webrick] How to extract filename from a request?

Daniel V�lkerts wrote:

> 
> This is my flow
> 
> Browser is pointed to http://localhost/filter/filter1.fd
> 
> As my Servlet ist mounted to /filter he regonize the request and thats 
> the part I'd like to implement -> now the servlet should look up the 
> filter to use (here filter1.fd) and should parse this filter file for 
> directives which controll the output.

So, the part of the URL that comes after /filter/ is data to be passed 
to some method?

> 
> So I've to extract the filename from the request_uri. The way I went 
> today is to call
> 
> filterName = req.request_uri.to_s.scan(/\w+\.fd/)
> 
> My question was if there is another possible better ruby way or variable 
> included with webrick to accomplish this task.


If you have /filter/ mapped to a servlet, and call 
http://localhost/filter/someFilterFile, then, in your servlet, 
request.path should give you "/filter/someFilterFile"

   def do_GET(req, res)
     url_parts = req.path.split( '/' )
     base = url_parts[0]
     filter_file_path = url_parts[1]
   end


That's one way to grab the data.  (Not the best; no error checking, etc.)

James