[Webkit-unassigned] [Bug 213502] [GTK][WPE] Change the cookies accept policy when ITP is enabled

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=213502

--- Comment #18 from Carlos Garcia Campos <cgarcia at igalia.com> ---
(In reply to Michael Catanzaro from comment #17)
> (In reply to Carlos Garcia Campos from comment #16)
> > Those are failing because the cookie is not set after being granted due to
> > the no-third-party accept policy (which is the default in tests for all
> > ports). Those tests don't change the ThirdPartyCookieBlockingMode, so
> > OnlyAccordingToPerDomainPolicy is used and we don't change the policy in
> > that case. 
> 
> I think we could still make this work by having the GTK API layer fudge the
> cookie policy (since the test controller will not run that code), rather
> than doing it in the network layer.

It doesn't matter, the storageAccess tests I mentioned would still fail, because no-third-party is the default, so setting the cookie after granted by storage access is rejected by the cookies jar. That's exactly what this bug is about.

> But maybe it's better to just document that ITP makes the most sense when
> used with ACCEPT_ALWAYS, maybe print a g_warning() if ITP is enabled and the
> cookie policy is NO_THIRD_PARTY, but let apps set it anyway.

Or maybe I don't understand the different ThirdPartyCookieBlockingMode, but I'm pretty sure we want to use All unconditionally, otherwise we would be less restrictive than before.

> > I don't know why we are enabling ITP in all tests either.
> 
> Doesn't matter IMO. Let's just match what Apple does.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20200703/e796d708/attachment-0001.htm>