[Webkit-unassigned] [Bug 103513] Provide the backend for exposing the layer tree to the Web Inspector

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Thu Nov 29 10:14:52 PST 2012


https://bugs.webkit.org/show_bug.cgi?id=103513





--- Comment #23 from Antoine Quint <graouts at apple.com>  2012-11-29 10:17:05 PST ---
(In reply to comment #22)
> > OK. The intent is that the front-end should be able to get from a layer to its DOM node for inspection. I understand it's expensive to push the node for each and every node with a RenderLayer. Do you think it would be appropriate to add a new method on this agent to get the DOM id for a given RenderLayer? That way the front-end would request the id as needed only.
> 
> Given that you maintain ids for your layers, adding protocol methods for resolving layer to a node and vice versa would make sense to me.

I'm adding a new "nodeIdForLayerId" method on the agent. There will likely be a need for the reverse, but I plan on addressing this in a future patch when it becomes a necessity.

> > I honestly don't have enough data to provide you with a payload size. Right now, the agent is only enabled once the developer asks for the layer tree to be displayed (ie. once .enable() is called). I'm not entirely sure what the lifecycle of the agent is currently, but should we make it so that it's disabled once the relevant piece of UI for it is not visible thus limiting the amount of information sent through this agent?
> 
> I guess you could start as is and add greater granularity as you go. An option would be to send "layerTreeUpdated" invalidation signal with no parameters and make client re-request the tree on demand. So that you did not load the protocol while say the tree is collapsed or offscreen.

That sounds like a good plan.

> > Are you suggesting I add a new static method on InspectorInstrumentation to allow communication between WebCore and the agent? If so, I will do this, thanks for pointing it out.
> 
> Yep. You should ack like the other agents.

I got this working along with fixes for all your other comments, I plan on sending a revised patch for review later today.

-- 
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