Re: Questions on Gtk::Notebook

[Antonio Coralles <noche suapie reflex at>]

Antonio Coralles wrote:

> Murray Cumming wrote:
>
> > On Wed, 2005-07-20 at 03:21 +0200, Antonio Coralles wrote:
> > > Murray Cumming wrote:
> > >
> > > > On Tue, 2005-07-19 at 22:55 +0200, Antonio Coralles wrote:
> > > >>
> > > >> plain text document attachment (NOTEBOOK.txt)
> > > >> After deriving my own Gtk::Notebook class, i experienced the 
> following
> > > >> behavour, for which i want to know, if this is on purpose:
> > > >>
> > > >> 1.) A Gtk::Notebook seems to emit signal_switch_page with a second
> > > > parameter 0
> > > >>     when realized. Because i wanted to open a different page by
> > > > default, i had
> > > >>     to override on_realize().
> > > >>
> > > >> 2.) A Gtk::Notebook emits the same signal a few times, when getting
> > > > destroyed.
> > > >>     This led to following warnings glibmm-warnings:
> > > >>
> > > >>     (duty-roaster:20667): glibmm-WARNING **: 
> Glib::create_new_wrapper:
> > > >>     Attempted to create a 2nd C++ wrapper for a C instance whose C++
> > > > wrapper
> > > >>     has been deleted.
> > > >
> > > > If you can make a test case for this, please add it to bugzilla.
> > >
> > > Did you mean a testcase for the above glibmm warning, or for the fact
> > > the signal_switch_page is emited on destruction ?
> >
> > For the warning. It's obviously very strange and wrong.
>
> Ok, creating a test case for this warning is not as easy as i had
> thought, but i'm working on it ...

Hmm, allthough i tried to reproduce that warning with a small test 
class, i was not able to do this, at least for now. The only way to 
reproduce it at the moment is:
-) download duty-roaster-0.0.79.94
-) open /src/interface/shifttable/yearsview.cc
-) comment or remove line 28, inside YearsView::~YearsView()
-) compile, and start the program.
-) close it and you'll get the warning.
Sorry, that i didn't found a more elegant way ...
Antonio