isDisabledFormControl() should apply for shadow dom element also

[Santosh Mahto]

Currently isDisabledFormControl() check for all form control element whether it is disabled or not
so for <select disable> </select> it will return true.

But is doesnot  apply to any shadow DOM element that are part of form control element 

 <Input disable>  </input>  --> isDisabledFormControl()  return true;
 but for  all shadow Nodes of input element  (TextControlInnerTextElement, SpinButtonElement)   it return false.

So If ShadowHost Node (form control) is disabled then isDisabledFormControl()  should return true for Host and also for Shadow Nodes.


consequence of this is that  shadow Nodes of Form control still receives mouse event while mouse event for Form Control element  is disabled

[Santosh Mahto]

Created attachment 202971 [details]
Patch

[Build Bot]

Comment on attachment 202971 [details]
Patch

Attachment 202971 [details] did not pass mac-wk2-ews (mac-wk2):
Output: http://webkit-queues.appspot.com/results/725120

New failing tests:
fast/forms/input-readonly-select.html
fast/forms/disabled-mousedown-event.html

[Build Bot]

Created attachment 202995 [details]
Archive of layout-test-results from webkit-ews-12 for mac-mountainlion-wk2

The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews.
Bot: webkit-ews-12  Port: mac-mountainlion-wk2  Platform: Mac OS X 10.8.3

[Benjamin Poulain]

Comment on attachment 202971 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=202971&action=review

Your patch needs a test, a description in the ChangeLog.
It looks like it also broke some tests. Make sure you run the tests before submitting something.

> Source/WebCore/dom/Element.cpp:1142
> +    Element* host = shadowHost();
> +    if (host)
> +        return host->isDisabledFormControl();

Following WebKit's style, this would be: 
if (Element* host = shadowHost())
    return host->isDisabledFormControl();

[Benjamin Poulain]

Until we re-enable web components, shadow dom is pretty much internal.
It would be good to explain why this change is needed.

I guess a shadow tree is visible for a form element, the state "disabled" changes from JavaScript, and the tree above is still usable?

[Santosh Mahto]

(In reply to comment #5)
> Until we re-enable web components, shadow dom is pretty much internal.
> It would be good to explain why this change is needed.
    the change is not final, first i  need review comments to validate the points.

> I guess a shadow tree is visible for a form element, the state "disabled" changes from JavaScript, and the tree above is still usable? 
   
the bug  already here that is related to this is https://bugs.webkit.org/show_bug.cgi?id=102841, 
even I am not sure of exact behaviour but I think that if form element  is disabled then all of associated  shadow dom nodes should not receive mouse event

In webkit we dont dispatch mouse event to node if it is form element and disabled. I think the behaviour should extend for shadow dom node.

I need  more views on my points.
adding appropriate guy would certainly help in deciding 

and the tree above is still usable?--> No  because host>isDisabledFormControl();


Sorry for test failure as I am not mac man