503078 - content-location should be used as the base URI

Status: VERIFIED DUPLICATE of bug 109553

(Core :: Networking: HTTP, defect)

Description

[Pierre-Antoine Champin]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; fr; rv:1.9.1) Gecko/20090624 Firefox/3.5
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; fr; rv:1.9.1) Gecko/20090624 Firefox/3.5

According to RFC 2616 (HTTP/1.1), the value of the Content-Location HTTP field should be used as the base URI instead of the requested URI.
see http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.14

Firefox does not comply with that, as illustrated with the URL provided with this bug
http://liris.cnrs.fr/~pchampin/tmp/content-location-test/

Reproducible: Always

Steps to Reproduce:
1. Go to http://liris.cnrs.fr/~pchampin/tmp/content-location-test/
Actual Results:  
The link "see the result" points to http://liris.cnrs.fr/~pchampin/tmp/content-location-test/result.html

Expected Results:  
It should point to http://liris.cnrs.fr/~pchampin/tmp/content-location-test/success/result.html (using the Content-Location URI as a base URI for the link).

Tested on Opera and Safari as well, and they do not pass the test :-(

Comment 2

[Boris Zbarsky [:bzbarsky]]

Verified.  Can't do this, because far too many servers send bogus Content-Location due to the spec being so vague.

Comment 3

[Pierre-Antoine Champin]

I skimmed the dublicate bug and it dependancies; I think I understand the arguments backing this decision, but I still do not see how the spec is vague. It is quite clear to me.

However, I understand the tension for Firefox between strict standard compliance with the standard, and "compliance" with the actual web. It is a shame though, because I think Content-Location has interesting use-cases...