[PD] loadbang behavior

[Miller Puckette]

Oh, I see :)

the "loadbang" method is now called with a floating-point argument, so you
can do this:

static void loadbang_loadbang(t_loadbang *x, t_floatarg action)
{
    if (action == LB_LOAD)
        outlet_bang(x->x_obj.ob_outlet);
}

and just get the one bang.  This was done to allow "init" and "Close" bangs
as well.

cheers
Miller

On Mon, Jan 02, 2017 at 05:34:02PM +0100, Antoine Villeret wrote:
> here is a sample code that demonstrate the issue :
> https://raw.githubusercontent.com/avilleret/helloworld/loadbang_issue/helloworld.c
> 
> When the object is created, loadbang method is called.
> Save the patch.
> Close it
> Reopen it, then the loadbang method is called twice (maybe one time at
> object loading and one more at patcher loadbang).
> 
> 
> 
> --
> do it yourself
> http://antoine.villeret.free.fr
> 
> 2017-01-02 17:25 GMT+01:00 Miller Puckette <msp at ucsd.edu>:
> 
> > Happy new year to you...
> >
> > I don't get this behavior - can you send a sample patch?
> >
> > thanks
> > Miller
> >
> > On Mon, Jan 02, 2017 at 04:50:47PM +0100, Antoine Villeret wrote:
> > > Hello,
> > >
> > > Happy new year pd-list !
> > >
> > > With pd-0.46-7, loadbang method was called only once when patcher loads.
> > > Now with 0.47-1, loadbang method are now called twice.
> > > Why loadbang method is called twice ?
> > >
> > > As far as I understand, loadbang method could be called not only on
> > patcher
> > > load (LB_LOAD), but also when closing an abstraction (LB_CLOSE) and
> > before
> > > the object is connected to parent patcher (LB_INIT).
> > >
> > > Many externals uses this loadbang method to initiate their state.
> > > This new behavior trig a lots of "wrong argument for loadbang method"
> > error
> > > in pd console.
> > >
> > > Is it a good practice to rely on loadbang call to initiate objects ? or
> > > should we avoid it and use [loadbang] object and/or make initialisation
> > in
> > > object_new method ?
> > > --
> > > do it yourself
> > > http://antoine.villeret.free.fr
> >
> > > _______________________________________________
> > > Pd-list at lists.iem.at mailing list
> > > UNSUBSCRIBE and account-management -> https://lists.puredata.info/
> > listinfo/pd-list
> >
> >