Re: GtkExpander/GtkDisclosure

[Ray Strode <halfline hawaii rr com>]

I just wanted to say, first, that I don't think your code is really all 
that complex anyway,
so if I were you I wouldn't go through the work to change it.

But, if you wanted to ...

> 	Yes, I considered this. But what you'd really end with is both
> approaches. You'd need to implement a GtkCheckButton sub-class and embed
> that in the GtkBin sub-class. Also, it would be a little weird because
> most of the operations you need to do would be on the GtkCheckButton
> sub-class and that's where the "toggled" signal would come from too.
I don't think you would need to sub-class GtkCheckButton.  You could 
just use its public API for what you need I think.  You're right though, 
you would end up having to relay some of the stuff to your widget.

> 	So, yes I think it would simplify the implementation but I think it
> would be a very confusing API for such a conceptually simple widget.
I think either way could use the same external API.

> 	You need to recieve button press and motion/crossing events. You can't
> do that with a NO_WINDOW widget right ?
>  
I was actually talking about if you switched to using GtkCheckButton. 
Note though that a NO_WINDOW widget can still have a GdkWindow in order 
to receive events.  GtkButton is an example of this, I think.  I think 
doing that for your widget would make the code more complex though.

In the end, I don't think implementation matters too much, as long as 
the API is good and the widget isn't noticably slow.

--Ray