- From: Yosi Scharf <syosi@mit.edu>
- Date: Tue, 04 Jan 2005 18:47:18 -0500
- To: Dan Connolly <connolly@w3.org>
- Cc: public-cwm-bugs@w3.org
Dan Connolly wrote: >In debugging his photo-calendar, timbl ran accross some >URIs with spaces. I did this test: > >$ echo "<a b> <c> <d>." | python ~/w3ccvs/WWW/2000/10/swap/cwm.py > <a b> <c> <d> . > >and tim said that's not what he'd like to see. > >So let's call it a bug. > > > Can this be clarified? What is the desired behavior? Cwm is following n3.n3, which defines an explicit uri as "<[^>]*>" So is this bug a statement that that line in n3.n3 should disallow some characters that clearly cannot show up in uri's, and cwm should then follow n3.n3? A quick read of http://www.faqs.org/rfcs/rfc2396.html , section 2.4.3, gave me something like "<[^\u0000-\u001F\u007F <>#%\\\"]*>" for the regexp for a uri. Yosi
Received on Tuesday, 4 January 2005 23:47:51 UTC