Re: precalc_first

On Sun, Apr 18, 2021 at 01:10:35PM -0400, al davis wrote:
> On Sun, 18 Apr 2021 12:40:51 +0200
> Felix Salfelder <[email protected]> wrote:
> >  I have removed
> > the call from list_save and recorded the changes in the prec_list
> > branch. It has the opposite effect to forwarding precalc_first in
> > subcircuit proto, see precalc_first branch.
> 
> This demonstrates the need for the call to precalc_first in list_save.

So, don't we need something like

void DEV_SUBCKT_PROTO::precalc_first()
{
  BASE_SUBCKT::precalc_first(); // skip DEV_SUBCKT
  assert(subckt());
  subckt()->precalc_first();
}

as I tried in the precalc_first branch?

I have added another test that lists a switch model contained in a
subcircuit -- it also fails without the above.

(I still don't think that DEV_SUBCKT_PROTO::precalc_last needs changing,
contrary to what you said. There might be some misunderstanding
involved?)

thanks
felix