Re: Gtk::ListStore bug?

[Paul Davis <pjdavis engineering uiowa edu>]

Murray Cumming wrote:

> On Sun, 2006-04-30 at 02:10 -0500, Paul Davis wrote:
>  
>
> > Hey,
> >
> > I'm attempting to get a signal from a list store when the rows in the 
> > model change.  But it doesn't seem to be emitting the signal.
> >
> > I went and looked at the source and I don't see that ListStore is 
> > calling rows_reordered like its supposed to.
> >
> > From the docs:
> >
> > void Gtk::TreeModel::rows_reordered...
> >
> > Emits the "row_reordered" signal on tree_model.
> > This should be called by models when their rows have been reordered.
> >
> > And in liststore.ccg, there's no reference to rows_reordered.
> >    
>
> It's a wrapper for C code.
>
>  
I noticed after a bit that alot of the functionality seemed to pass 
through.  I had (incorrectly) assumed that the sigc stuff would have to 
be coded ontop of Gtk.  Silly me.

> >  Granted I 
> > haven't gone through and extensive trace of the class hierarchy, but 
> > doing some grepping I don't see where it should be getting called.
> >
> > If anyone knows a better way to detect when the user rearanges the rows 
> > in a liststore, I'd love to hear about it.
> >    
>
> Do you mean via drag and drop?
>
>  
I mean by using the set_reorderable() method.  I don't do anything to 
override the default reorderable functionality.  I did figure out that 
the signal_row_deleted is being called when a reorder event occurs, so I 
just used it as a hack.  So for now, its working, but its one of those 
things I'd rather use the right signal if possible.

Thanks,
Paul Davis