Bug 145339 – Rationalize the behavior when timing out touch event dispatch

• Status: NEW
• Component: New Bugs

[Benjamin Poulain]

Rationalize the behavior when timing out touch event dispatch

[Benjamin Poulain]

Created attachment 253636 [details]
Patch

[Sam Weinig]

Comment on attachment 253636 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=253636&action=review

> Source/WebKit2/Shared/WebEvent.h:415
> +    bool hasEventDeliveryTimedOut() const { return m_eventDeliveryTimedOut; }
> +    void eventDeliveryTimedOut() { m_eventDeliveryTimedOut = true; }

I've been trying to keep these platform Event objects immutable so far. Is there another design that doesn't force mutability here?

[Benjamin Poulain]

(In reply to comment #2)
> Comment on attachment 253636 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=253636&action=review
> 
> > Source/WebKit2/Shared/WebEvent.h:415
> > +    bool hasEventDeliveryTimedOut() const { return m_eventDeliveryTimedOut; }
> > +    void eventDeliveryTimedOut() { m_eventDeliveryTimedOut = true; }
> 
> I've been trying to keep these platform Event objects immutable so far. Is
> there another design that doesn't force mutability here?

Sure. I don't have to put that information on the event, it can be an additional structure passed along to the WebProcess.

Would that be okay with you?

[Sam Weinig]

(In reply to comment #3)
> (In reply to comment #2)
> > Comment on attachment 253636 [details]
> > Patch
> > 
> > View in context:
> > https://bugs.webkit.org/attachment.cgi?id=253636&action=review
> > 
> > > Source/WebKit2/Shared/WebEvent.h:415
> > > +    bool hasEventDeliveryTimedOut() const { return m_eventDeliveryTimedOut; }
> > > +    void eventDeliveryTimedOut() { m_eventDeliveryTimedOut = true; }
> > 
> > I've been trying to keep these platform Event objects immutable so far. Is
> > there another design that doesn't force mutability here?
> 
> Sure. I don't have to put that information on the event, it can be an
> additional structure passed along to the WebProcess.
> 
> Would that be okay with you?

Yeah, that sounds great!

[Brent Fulgham]

Comment on attachment 253636 [details]
Patch

It sounds like Sam wants you to make changes to the implementation, so I'll r- this to get it our of the review queue.

[Simon Fraser (smfr)]

rdar://problem/27666110