[Webkit-unassigned] [Bug 68610] Microdata: Basic implementation of document.getItems() method

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=68610





--- Comment #17 from Arko Saha <nghq36 at motorola.com>  2011-10-12 00:44:38 PST ---
(In reply to comment #13)
> Are you trying to detect a case where the script didn't pass type? I don't think comparing it to the string 'undefined' is the right approach.

Ok, we will use ConvertUndefinedOrNullToNullString which converts undefined to null string.

> Please use OVERRIDE macro.
> 
I am not aware of this macro. Could you please help me to understand this.

> You also need to modify NodeListsNodeData::isEmpty.
> 
Ok, we will modify this.

> Please modify invalidateCachesThatDependOnAttributes instead. m_classNodeListCache is a good one to mimic. r- because of this.

In case of class attribute when classAttr changes it calls classAttributeChanged() from StyledElement::parseMappedAttribute().
Now classAttributeChanged() invokes dispatchSubtreeModifiedEvent() which internally calls invalidateCachesThatDependOnAttributes() to invalidate m_classNodeListCache.

Earlier I did the same for itemtype attribute change. We were calling dispatchSubtreeModifiedEvent() from itemTypeAttributeChanged()
which internally invoked invalidateCachesThatDependOnAttributes() to invalidate m_microDataItemListCache.

Now we modified the code so that it will only invalidate the m_microDataItemListCache when itemType attribute changes, it should not send the dispatchSubtreeModifiedEvent() as you suggested.
Do we need to follow the same as earlier case?

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