Re: parsing URI (references) according to RFC 3986

On 6/23/11 2:46 PM, Julian Reschke wrote:
>> There is no question about those aspects of this. The base URI for the
>> image source in this case is
>> "http://greenbytes.de/tech/tc/uris/imgother.html".
>
> Yes.
>
> So please remind me (sorry; has been a long day for me): what is the
> problem here?
>
> Sending an additional request despite the URI spec says "should not"?
> Why is the request being sent anyway?

The problem is that per 4.4 as I understand it this HTML:

   <!DOCTYPE html>
   <base href="http://greenbytes.de/tech/tc/uris/imgother.html">
   <img src="#foo">

located at <http://greenbytes.de/tech/tc/uris/img.html> should treat the 
image load as a load of <http://greenbytes.de/tech/tc/uris/img.html> 
(because this is a same-document URI reference) whereas this is not what 
any browser does.  Browsers treat it as a load of 
<http://greenbytes.de/tech/tc/uris/imgother.html>.

Is my understanding of 4.4 incorrect?  If so, why?

>> If it's not supposed to mean anything and specs that use URIs are just
>> supposed to define what happens (a stance I agree with), then why is 4.4
>> even there?
>
> The text was changed from RFC 2396, which just said:

Yes. I'm aware of what RFC 2396 said.  The substantive changes from RFC 
2396 to RFC 3986 are that the bit about empty URIs in contexts expected 
to result in a new request was removed and the definition of a 
same-document reference was changed to be defined in terms of comparison 
of a resolved URI to a base URI instead of examination of the original 
URI reference string.

My example above does not involve either of those changes, so it's 
equally an issue using the RFC 2396 wording.

> The history for this seems to be in
> <http://labs.apache.org/webarch/uri/rev-2002/issues.html#017-rdf-fragment>.

To be clear, I'm not so much asking why the text is in RFC 3986 or RFC 
2396.  I understand the problem it was attempting to solve.  I'm just 
saying that:

1)  It's the wrong way to solve the problem.
2)  The problem is outside the remit what I think a URI RFC should be
     concerned with.
3)  And most importantly, it's a partial answer to the original "Which
     parts of RFC 3986 are interop issues?" question.

My primary interest in posting here at all was to bring up item 3 above, 
since you _specifically_ asked for a list of such interop issues.

What you decide to do from this point on is up to you, but I feel like 
I'm seriously wasting my time here....

-Boris

Received on Thursday, 23 June 2011 18:56:55 UTC