typematic surpresses events unnecessarily

[bill]

Starting from the initial checkin in [7522], typematic stops propagation for a number of events.

The current code is:

on(node, "mousedown", lang.hitch(this, function(evt){
        event.stop(evt);
        typematic.trigger(evt, _this, node, callback, node, subsequentDelay, initialDelay, minDelay);
})),
on(node, "mouseup", lang.hitch(this, function(evt){
        if(this._obj){
                event.stop(evt);
        }
        typematic.stop();
})),
on(node, "mouseout", lang.hitch(this, function(evt){
        event.stop(evt);
        typematic.stop();
})),
on(node, "mousemove", lang.hitch(this, function(evt){
        evt.preventDefault();
})),
on(node, "dblclick", lang.hitch(this, function(evt){
        event.stop(evt);
        if(has("ie")){
                typematic.trigger(evt, _this, node, callback, node, subsequentDelay, initialDelay, minDelay);
                setTimeout(lang.hitch(this, typematic.stop), 50);
        }
}))

I didn't test it, but seems like at a minimum the event.stop() calls will interfere with widgets or apps that are trying to monitor these events at a higher level, for example:

<div data-dojo-type=dijit.form.NumberSpinner onMouseMove="...">

(mouse movement over the up/down arrows won't be reported, will it?)

I also don't see a reason for the code to be there. In particular the mousemove handling seems strange since mouse movement doesn't have anything to do with typematic, does it?

Compare to [27394] where I had to add back the propagation of mousedown.

[Douglas Hays]

In [27994]:

Fixes #14743. Change event.stop to preventDefault to allow parental monitoring. Remove mousemove intercepts since this doesn't seem to affect typematic behavior.