Closed GoogleCodeExporter closed 8 years ago
I've discovered that this issue does not seem to occur in IE 7 (I was using IE
8
before).
Original comment by phil.boo...@gtempaccount.com
on 27 Nov 2009 at 1:14
Having looked at the code, it seems that the decoration of header's children is
an
interesting area. In Firefox and IE, it is postponed until the entire documents
is
ready, and then applied with the aid of the CSS class 'header-ready'. I'm not
yet
aware of why it has to be done like this.
Anyway, a consequence of this is that the test will actually pass in Firefox
HTML
mode, provided you allow 'sufficient' time for the decoration to occur. In
terms of
automating tests in Selenium, this means waiting until some time after
xforms-ready
has been dispatched.
Obviously this is in no way a solution in and of itself, but I thought it was
worth
noting.
Original comment by phil.boo...@gtempaccount.com
on 27 Nov 2009 at 1:21
I never realised until this morning, hence the previous comment, but header
supports
@nodeset. So decoration has to be postponed until the iterations can be worked
out
druing rewire().
Original comment by phil.boo...@gtempaccount.com
on 30 Nov 2009 at 2:29
A fix is awaiting review at r3201.
The fix itself is somewhat hackish. In the problem cases, the call to
UX.addClassName
in Header.onDocumentReady() is not working (decoration of child elements does
not
occur). Yet, if the call is made in the constructor, it correctly invoked child
decoration. I've no idea why that is the case, but it is.
The other part of the fix is to add a binding rule for header to the CSS, so
that it
gets decorated (again, it wasn't until now in the problem cases).
Original comment by phil.boo...@gtempaccount.com
on 30 Nov 2009 at 3:52
Original comment by phil.boo...@gtempaccount.com
on 30 Nov 2009 at 3:57
In trunk at r3205.
Original comment by phil.boo...@gtempaccount.com
on 2 Dec 2009 at 11:57
Original issue reported on code.google.com by
phil.boo...@gtempaccount.com
on 24 Nov 2009 at 11:38Attachments: