[Webkit-unassigned] [Bug 26718] [Gtk] Add support for javascript windows for DRT

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=26718





------- Comment #9 from jmalonzo at gmail.com  2009-06-30 18:02 PDT -------
(In reply to comment #6)
> (From update of attachment 31923 [review])
> >  void ChromeClient::closeWindowSoon()
> >  {
> > -    notImplemented();
> > +    // Clear the page group name (because mac does!)
> > +    webkit_web_view_set_group_name(m_webView, "");
> 
> Make this a FIXME. We still need to figure out how PageGroups will be useful,
> or not, for us.

I've added a FIXME here. I've added a comment where I've used this function in
DRT and why it's used there.

> 
> > +    webkit_web_view_stop_loading(m_webView);
> > +
> > +    g_signal_emit_by_name(m_webView, "close-web-view");
> >  }
> 
> I wonder if the signal should be emitted first, and give the application the
> oportunity of avoiding the closing of the window. I think this was discussed
> already?

I've added my own reasoning in comment #5. But with the updated patch, I've
updated this to let the application handle the signal.

> 
> > +    settings->setJavaScriptCanOpenWindowsAutomatically(javascriptCanOpenWindows);
> 
> The meaning of this setting is a bit nebulous to me. Without turning it on we
> were able to let javascript create new windows, so why do we need it now?

As discussed in IRC, this is for opening windows that aren't initiated by a
user action.

> 
> > -void LayoutTestController::setPopupBlockingEnabled(bool popupBlockingEnabled)
> > +void LayoutTestController::setPopupBlockingEnabled(bool flag)
> >  {
> > -    // FIXME: implement
> > +    WebKitWebView* view = webkit_web_frame_get_web_view(mainFrame);
> > +    ASSERT(view);
> > +
> > +    WebKitWebSettings* settings = webkit_web_view_get_settings(view);
> > +    g_object_set(G_OBJECT(settings), "javascript-can-open-windows", flag, NULL);
> > +
> >  }
> 
> hrm; this looks very wrong. Enabling popup blocking by setting
> javascript-can-open-windows to TRUE? I'm under the impression that there is a
> misunderstanding about the meaning of this setting. I'm going to say r-.

flag should've been !flag. Fixed.

Thanks.


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