I am starting to think this is not a bug ;)
process = sin(0:!);
is equivalent to
process = sin;
because 0:! has no inputs/outputs. similarly,
process = sin(!);
acts as
process = !, sin;
So I think I understand the logic now, but it was not obvious to me and
I would like to hear your comments.
Thanks,
Oleg.
On 03/31, Oleg Nesterov wrote:
>
> OK, this is not unique to select2 ...
>
> Say,
> process = sin(!);
>
> compiles to
>
> output0[i0] = FAUSTFLOAT(std::sin(float(input1[i0])));
>
> Or,
> process = atan2(1,!);
>
> results in
>
> output0[i0] = FAUSTFLOAT(std::atan2(1.0f, float(input1[i0])));
>
> and please note that
>
> process = atan2(!,1);
>
> generates the same code
>
> output0[i0] = FAUSTFLOAT(std::atan2(1.0f, float(input1[i0])));
>
> I feel this has something to do with boxPrimX() logic, but I am not
> sure...
>
> Oleg.
>
> On 03/31, Oleg Nesterov wrote:
> >
> > Hello,
> >
> > I think I've accidentally run into a problem I can't explain.
> >
> > Consider this trivial code:
> >
> > process = select2(1, 0, !);
> >
> > I agree, it makes no sense, but faust happily compiles it to
> >
> > void compute(int count, FAUSTFLOAT** inputs, FAUSTFLOAT** outputs)
> > {
> > FAUSTFLOAT* input0 = inputs[0];
> > FAUSTFLOAT* input1 = inputs[1];
> > FAUSTFLOAT* output0 = outputs[0];
> > for (int i0 = 0; i0 < count; i0 = i0 + 1) {
> > output0[i0] = FAUSTFLOAT(float(input1[i0]));
> > }
> > }
> >
> > and this doesn't look right no matter what. It seems that "!" inside select2
> > always has 2 inputs and acts as
> >
> > cut(x,y) = y;
> >
> > I can probably understand this to some extent, "!" should cut an input and
> > at the same time select2() should have an output, but to me this code should
> > not compile.
> >
> > Can you explain this?
> >
> > Oleg.
_______________________________________________
Faudiostream-devel mailing list
Faudiostream-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/faudiostream-devel
