chicken-janitors
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Chicken-janitors] Re: #396: Handle empty path components in uri-generic


From: Chicken Trac
Subject: [Chicken-janitors] Re: #396: Handle empty path components in uri-generic/uri-common
Date: Wed, 22 Sep 2010 13:14:05 -0000

#396: Handle empty path components in uri-generic/uri-common
-------------------------+--------------------------------------------------
  Reporter:  sjamaan     |       Owner:  sjamaan                                
 
      Type:  defect      |      Status:  new                                    
 
  Priority:  minor       |   Milestone:  4.7.0                                  
 
 Component:  extensions  |     Version:  4.6.x                                  
 
Resolution:              |    Keywords:  uri, pedantic strictness, rfc, 
standards
-------------------------+--------------------------------------------------

Comment(by sjamaan):

 Replying to [comment:4 iraikov]:
 >
 > I don't think you are complaining, I just wondered where this issue is
 addressed in the RFC, since the ticket description refers to a "quick
 check of the spec", and there is no test case that illustrates a specific
 problem. It is perfectly possible that this behavior causes some breakage
 somewhere, especially if it is undefined in the RFC.

 Which behaviour do you mean?

 I just spent an hour implementing a nasty workaround to encode empty path
 components for a project at work because Apache discards empty path
 components. It would be nice if our implementation kept them around. I've
 also seen bugreports elsewhere that browsers keep them around so a
 relative reference generated on a server that's unaware of the extra slash
 breaks on the browser because it does understand the extra slash (and ..
 would strip off one extra slash).

 I think discarding empty components causes more trouble than keeping them.
 I'm sure keeping them would also cause trouble with some other
 applications, but only if they generated wrong URLs in the first place.
 The user of the uri lib can always decide to discard them (Spiffy's file
 handler could do this, for example so that requests for files in the
 filesystem would work with any number of slashes, but it could keep around
 the empty strings in the pathinfo part so web applications can cleanly
 represent any string there without jumping through hoops like I had to do
 with Apache)

-- 
Ticket URL: <http://bugs.call-cc.org/ticket/396#comment:5>
Chicken Scheme <http://www.call-with-current-continuation.org/>
Chicken Scheme is a compiler for the Scheme programming language.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]