Avoid heap allocation for EventContexts

[Ryosuke Niwa]

We don't need to a class hierarchy to represent different kinds of EventContext.
Merge all the subclasses of EventContext onto itself to avoid heap allocation in EventPath.

[Ryosuke Niwa]

Created attachment 420804 [details]
WIP

[Ryosuke Niwa]

Created attachment 420901 [details]
WIP2

[Ryosuke Niwa]

Created attachment 421066 [details]
WIP2 + cleanup

[Radar WebKit Bug Importer]

<rdar://problem/74586915>

[Ryosuke Niwa]

Created attachment 421500 [details]
Patch

[Simon Fraser (smfr)]

Comment on attachment 421500 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=421500&action=review

> Source/WebCore/dom/EventContext.h:62
> +    bool isMouseOrFocusEventContext() const { return m_type == Type::MouseOrFocus; }
> +    bool isTouchEventContext() const { return m_type == Type::Touch; }
> +    bool isWindowContext() const { return m_type == Type::Window; }

Not this patch, but it's weird how the context is sometimes about the type of event, and sometimes about the object it's fired on?

> Source/WebCore/dom/EventContext.h:104
> +    , m_type(type)

Can't this be m_type { type } ?

[Ryosuke Niwa]

(In reply to Simon Fraser (smfr) from comment #6)
> Comment on attachment 421500 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=421500&action=review
> 
> > Source/WebCore/dom/EventContext.h:62
> > +    bool isMouseOrFocusEventContext() const { return m_type == Type::MouseOrFocus; }
> > +    bool isTouchEventContext() const { return m_type == Type::Touch; }
> > +    bool isWindowContext() const { return m_type == Type::Window; }
> 
> Not this patch, but it's weird how the context is sometimes about the type
> of event, and sometimes about the object it's fired on?

Yup, we should do that. I was initially going to do that in this patch but I decided against it to keep this patch simple & mechanical.

> > Source/WebCore/dom/EventContext.h:104
> > +    , m_type(type)
> 
> Can't this be m_type { type } ?

Oops, fixed. Not sure what happened there.

Thanks for the reviews!

[Ryosuke Niwa]

Committed r273477 (234556@main): <https://commits.webkit.org/234556@main>