- From: Roy T. Fielding <fielding@gbiv.com>
- Date: Tue, 6 Apr 2004 13:12:12 -0700
- To: Graham Klyne <GK@ninebynine.org>
- Cc: uri@w3.org
> In appendix A, collected syntax, the production for absolute-URI > differs from that given in the body text. Yep, I missed that -- I wish the collected ABNF could just be generated from the XSLT. > Assuming that the body text production (using hier-part) is correct > (which I think it is) then the 'path' production is itself not used > anywhere, hence redundant. You are missing the whole point of having it. The text describing the syntax says their are five possible matching rules for a path, only one of which will be matched in any URI reference. That is why path is a top node and not used by the URI syntax itself. It also explains why those rules are not ambiguous, and hence why reducing them to one production is not desirable. Nevertheless, a non-validating grammar will combine all of those into one component that is called path. ....Roy
Received on Tuesday, 6 April 2004 16:14:18 UTC