[Webkit-unassigned] [Bug 34449] [Gtk] Evaluate and create tests for all the AtkRole's implemented by WebKitGtk

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Tue Feb 2 01:17:59 PST 2010


https://bugs.webkit.org/show_bug.cgi?id=34449





--- Comment #11 from Holger Freyther <zecke at selfish.org>  2010-02-02 01:17:59 PST ---
(In reply to comment #10)
> (In reply to comment #9)
> > I fear the unit tests are the wrong way. You should implement
> > WebKitTools/DumpRenderTree/gtk/Accessibility* for atk and then unskip the
> > Accessibility tests from LayoutTests/platform/gtk/Skipped.
> 
> I don't agree. There might be some overlap between DRT and the unit tests, but
> DRT won't exercise thoroughly the whole ATK APIs, which is what applications
> will be using.


> IMHO unit tests and DRT is an AND topic, not an OR one, but I guess I can be
> wrong.

My understanding is that we create a unit test when the specific thing can not
be tested with a LayoutTest and extending the LayoutTestController or its child
objects is too complicated.

In this case we have 73 Accessibility Layout Tests waiting for us. From a quick
glance they are testing the basics that will be tested with the Gtk+ specific
unit
tests. E.g. the Layout Tests should gurantee that "#1 of Joanmarie" is not
happening ever again.

So personally I would have loved to see the implementation of the DRT methods
first and then cover everything with unit tests. Well, Joanmarie has invested
quite some time in the ATK impl and the test cases so asking for them to be
thrown away is obviously not nice as well.

Is it asked too much if we would work on the DRT support first and then land
the test cases?

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



More information about the webkit-unassigned mailing list