[Webkit-unassigned] [Bug 47609] [Qt] Fix http/tests/misc/redirect-to-external-url.html

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=47609





--- Comment #7 from Jakub Wieczorek <jwieczorek at webkit.org>  2010-10-14 13:18:55 PST ---
(In reply to comment #6)
> (In reply to comment #2)
> > > [Qt] Fix http/tests/misc/redirect-to-external-url.html
> > 
> > Nah, you're not fixing it, really.
> > I recommend looking at what other ports are doing but special-casing a specific situation from a test case inside WebKit is not a solution at all. Not to mention that theoretically each scheme can be supported by the embedder via an extended QNAM.
> 
> Chromium special-cases it. My understanding is that the test is for WebCore functionality, PolicyChecker::continueAfterNavigationPolicy(PolicyAction policy and the canHandleRequest() is expected to fail.

OK. I don't think you should follow Chromium in this case though. Other ports don't seem to do this and yet, they pass this test, so apparently they can get the necessary information from the network stack. QtWebKit, due to its abstract network interface, cannot.

Moreover, if a specific behavior of canHandleRequest() is an assumption in this test case and yet that behavior never occurs in QtWebKit, what's the point of testing this anyway?

I think all this test deserves is a good comment in the skipped list.

> I agree that it would be better to remove the DRT special-casing we have in FrameLoaderClientQt, along with all the other DRT specific messages in there. But that's a different problem.

Yes. I'd suggest we don't make the situation even worse though.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.