[Webkit-unassigned] [Bug 83030] [EFL] DRT should support LayoutTestController's dumpFrameLoadCallbacks()

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=83030


Christophe Dumez <christophe.dumez at intel.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #136426|0                           |1
        is obsolete|                            |
 Attachment #136939|                            |review?, commit-queue?
               Flag|                            |




--- Comment #10 from Christophe Dumez <christophe.dumez at intel.com>  2012-04-12 11:31:14 PST ---
Created an attachment (id=136939)
 --> (https://bugs.webkit.org/attachment.cgi?id=136939&action=review)
Proposed patch

The previous tests was causing timeouts in the following tests:
* http/tests/eventsource/existent-eventsource-status-error-iframe-crash.html
* http/tests/eventsource/existent-eventsource-status-error-iframe-crash.html
* http/tests/eventsource/non-existent-eventsource-status-error-iframe-crash.html
* http/tests/misc/location-replace-crossdomain.html
* http/tests/misc/cached-scripts.html
* http/tests/misc/onload-remove-iframe-crash-2.html
* fast/dom/HTMLDocument/frameless-location-bugzilla10837.html
* fast/dom/gc-6.html
* fast/dom/attr_dead_doc.html
* fast/dom/frame-loading-via-document-write.html
* fast/dom/Window/Location/window-shadow-location-using-js-object-with-toString.html
* fast/dom/Window/closure-access-after-navigation-iframe.html
* fast/dom/constructors-cached-navigate.html
* fast/dom/HTMLObjectElement/object-as-frame.html
* fast/dom/HTMLHeadElement/head-check.html
* fast/dom/NodeList/adoptNode-node-list-cache.html
* fast/dom/DOMImplementation/createDocument-with-used-doctype.html
* fast/dom/gc-7.html
* fast/css-generated-content/positioned-div-with-floating-after-content-crash.html
* fast/css/user-stylesheet-crash.html
* fast/css/stylesheet-candidate-nodes-crash.xhtml

I updated the "load,started" signal that is emitted on the ewk_view so that the ewk_frame which started loading is passed to the listeners.
This way I can get rid of the following hack in EFL's DRT:

-void DumpRenderTreeChrome::onLoadStarted(void*, Evas_Object* view, void*)
+void DumpRenderTreeChrome::onLoadStarted(void*, Evas_Object* view, void* eventInfo)
 {
-    // FIXME: we actually need the frame related to this event
-    Evas_Object* frame = ewk_view_frame_main_get(view);
+    Evas_Object* frame = static_cast<Evas_Object*>(eventInfo);

After this fix, I can no longer experience any timeouts in the tests.

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