[PD-dev] Sending list of floats out an outlet

[Krzysztof Czaja]

hi Ben,

Ben Bogart - FMPM/F1999 wrote:
...
 >         t_atom f_out[9];
 >         t_atom* myf = f_out;
 >         SETFLOAT(myf,f);
 >         SETFLOAT(myf+1,f+10);
 >         outlet_list(x->myobj.ob_outlet,&s_list,2,myf);

this is fine.  Be cautious though -- whenever it comes to
incrementing myf (e.g. inside of a loop), do so _outside_ of
a SETFLOAT invocation -- this is a macro.

 > where f is a floatarg to the method. Why is it that Guenter is using
 > (t_atom*)&a_out rather than a_out as the *argv pointer? I copied the same

redundant, but does not hurt...

 >>On Tue, 16 Jul 2002, Krzysztof Czaja wrote:
...
 >>>if 'x' is your object pointer, 'n' is the number of floats, and
 >>>you already store them in an array of atoms 'a', then the call
 >>>will be
 >>>
 >>>outlet_list(x, &s_list, n, a);

oops, make it outlet_list(((t_object*)x)->ob_outlet, &s_list, n, a);
or as you did, i.e. more explicit, but avoiding a cast...