[Webkit-unassigned] [Bug 42578] [Qt] DRT sideeffect revealed by r63657

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=42578





--- Comment #16 from Andras Becsi <abecsi at webkit.org>  2010-10-19 03:12:45 PST ---
After changing:
WebKitTools/DumpRenderTree/qt/DumpRenderTreeQt.cpp:509
+    // create our primary testing page/view.
+    if (isGraphicsBased()) {
+        m_page = new WebPage(static_cast<QWebView*>(m_mainView), this);
+        static_cast<QWebView*>(m_mainView)->setPage(m_page);
+    }

to

WebKitTools/DumpRenderTree/qt/DumpRenderTreeQt.cpp:509
+    // create our primary testing page/view.
+    if (isGraphicsBased()) {
+        m_page = new WebPage(static_cast<WebViewGraphicsBased*>(m_mainView), this);
+        static_cast<WebViewGraphicsBased*>(m_mainView)->setPage(m_page);
+    }

8 tests fail:

http/tests/navigation/error404-basic.html
http/tests/navigation/error404-goback.html
http/tests/navigation/error404-subframeload.html
http/tests/navigation/javascriptlink-frames.html
http/tests/navigation/postredirect-basic.html
http/tests/navigation/postredirect-frames.html
http/tests/navigation/postredirect-goback1.html
http/tests/security/mixedContent/insecure-css-in-main-frame.html

The navigation tests seem to have wrong expected results, the security test was flakey before, now it seems to fail always.

The culprit is that the whole testing session is more than 1 minute slower on our bot after this patch.

We should discuss whether this is feasible, or if there is eventually a better way to achieve the goal.
My opinion is that it should be worth it especially if there are some other tests in the Skipped list which become correct after this change.

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