* Sjoerd Visscher <[EMAIL PROTECTED]> [2005-07-19 01:25]: > A. Pagaltzis wrote: > >He is correct, Tim. The base URI means “the URL where this > >document was found,” not “the prefix for any enclosed relative > >links.” I don’t see how RFC3986 can be read any other way. > > I am correct ;), but your explanation is wrong. 5.1 says > clearly that the "Base URI embedded in content" precedes any > other way of establishing a base URI.
You misunderstood what I said. The point is that regardless of how the base URI is determined (whether it is embedded in content or otherwise), it *means* that the content it applies to was actually found at the base URI. It’s not simply any arbitrary old prefix defined for convenience. > >So it would be > > > > <feed xmlns='http://www.w3.org/2005/Atom' > > xml:base='http://www.tbray.org/ongoing/ongoing.atom' > > xml:lang='en-us'> > > <title>ongoing</title> > > <link href='./' /> > > <link rel='self' href='' /> > > ... > > <entry xml:base='When/200x/2005/07/14/Atom-1.0'> > > If you want it to be perfect xml:base on the entry would be the > URI of the Atom Entry Document. And the html base URI should go > on the xhtml div element. Just like you get when the entries > and the div elements are XIncluded. Yes – good point. It was actually bugging me. > >If we wanted to avoid this, we should have added language to > >the spec that atom:[EMAIL PROTECTED]'alternate'] should never be > >considered a same-document reference, maybe. > > If your XML package has support for data-typing, xml:base and > same-document references, then you could no longer do > > Node n = link.getAttribute("href").dereference(); > > because Atom would no longer follow standard behaviour. That is a drawback, yes. Though at at that point you know it’s an atom:[EMAIL PROTECTED]'alternate'], so you *could* handle it specially – if the package has a decent API, that shouldn’t be too much effort either. Worst case, you might need to work with different URI resolvers or some such. In any case, it was just an idle thought. Regards, -- Aristotle Pagaltzis // <http://plasmasturm.org/>
