Andre Poenitz wrote:
> On Mon, Dec 16, 2002 at 06:17:03PM +0000, John Levon wrote:
>> On Mon, Dec 16, 2002 at 04:46:02PM +0100, Alfredo Braunstein wrote:
>>
>> > opens simultanously both the reference menu and the float menu.
>>
>> I guess this will fix the release-critical similar bug listed on
>> bugzilla too. Good stuff.
>>
>> > The problem is in InsetText::lfunMouseRelease(...), it checks if an
>> > inset
>>
>> Andre needs to ack this, he was in there last.
>
> I don't really understand the logic there. Ideally the innermost inset
> would return with "DISPATCHED" and no further action should be taken by
> the outside world.
>
> However, I don't know how the edit call fits in that scheme, especially as
> this implies a hard wired connection between right click and edit().
>
> I think whatever it does should be done in the inset's localDispatch(),
> but the call has been there "since ever"...
>
>> > Inset::localDispatch() on, say, InsetButton, making it to catch
>> > LFUN_MOUSE_RELEASE and call edit() by itself?
>>
>> A button is not necessarily editable ....
>
> This could be taken care of by doing "nothing special" in the insets'
> localDispatch instead of providing an empty(?) edit() function.
Something like this? Or should we overload localDispatch on every single
clase inheriting from InsetButton?
Inset::RESULT InsetButton::localDispatch(FuncRequest const & ev)
{
switch (ev.action) {
case LFUN_MOUSE_RELEASE:
if(editable()) {
edit(ev.view(),ev.x,ev.y,ev.button());
return DISPATCHED;
} else
return UNDISPATCHED;
default:
return UNDISPATCHED;
}
}
Regards, Alfredo
