[Webkit-unassigned] [Bug 98950] Bad copy constructor in StyleRareNonInheritedData

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=98950





--- Comment #13 from Rik Cabanier <cabanier at adobe.com>  2012-10-11 15:08:36 PST ---
(In reply to comment #12)
> View in context: https://bugs.webkit.org/attachment.cgi?id=168282&action=review
> 
> Looks good! I think the test could be simplified.
> 
> > LayoutTests/platform/chromium/TestExpectations:110
> > +# Fails because chromium doesn't have blendmode
> 
> I think it is better to just say that Blend Mode is not enabled on Chromium yet :)

OK
Will do

> 
> > LayoutTests/transitions/blendmode-transitions.html:17
> > +      -webkit-transition-duration: 1s;
> > +      -webkit-transition-timing-function: linear;
> > +      -webkit-transition-property: -webkit-box-shadow, text-shadow;
> > +	  -webkit-blend-mode: difference;
> 
> Why not just have a simple test to check the computed style before and after the change? This looks like it could be a simple JS test instead. Anyway, using a transition seems a little too much :)

The value is only wrong DURING the transition  :-)
So, I have to check midway through.

> 
> > LayoutTests/transitions/blendmode-transitions.html:33
> > +      -webkit-box-shadow: 0 -20px 10px red, 0 20px 10px blue;
> > +      text-shadow: 0 -20px 10px red, 0 20px 10px blue;
> 
> Do you really need the shadows?

I need something to transition. Does it matter if it's a shadow?

> 
> > LayoutTests/transitions/resources/transition-test-helpers.js:166
> > +                case CSSPrimitiveValue.CSS_IDENT:
> 
> I think you are still not testing this one. You should pass in an arbitrary string and see what happens and have it as a bad parsing test (also a JS only test could handle this).

The result of getcomputedstyle is an CSS_IDENT.
I will get a JS error without it.

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