Re: [gtk-list] Re: Questions about event percolation
- From: ramiro netscape com (Ramiro Estrugo)
- To: gtk-list redhat com
- Subject: Re: [gtk-list] Re: Questions about event percolation
- Date: Wed, 14 Apr 1999 13:23:33 -0700
ok, ive looked at how event selection is being setup and i am ok with
the proposed solution.
Federico Mena Quintero wrote:
> This is correct as well. If the child window is not selecting for a
> particular type of event, X will look for a window up in the hierarchy
> that *is* taking that type of event. Gtk obviously gets it and emits
> the appropriate signal.
> As a side question... why does Mozilla care if it gets motion events
> even if the mouse is over a child window?
Im using motion events as an example. The principle applies to any
The fundamental problem is that mozilla has its own event system and
compositor. It also has its own XP widget system. This complicates
things because we need to differentiate events occurring either in a
child or the parent container.
For example, a windowless xp widget, will use the parent container to
detect X events and dispatch these to the mozilla event system.
Not so for native widgets which do their "own" dispatching without help
from the parent.
This applies to all events to be able to do all the magical things that
web browsers do.
] [Thread Prev