Re: testing

On Mon, Feb 09, 2015 at 02:00:02PM -0500, al davis wrote:
> So I made another test case that adds a node.  gnucap-uf then 
> gave wrong answers because of a mismatch because the subcircuit 
> wasn't remapped.  I have some ideas, but for now losing the 
> probes is less of a problem than the alternatives.

thanks for the analysis. the frozen flag, that i forgot about. this once
amended a related issue. but sure, it's just a hack :(. i'll look into
your new test some time.

> Not really surprising.  autotools does a lot, all mixed up, in a 
> very complex way that is impossible to test adequately.

yes. it would still be good to have the useful fraction of this lot on
our side. i don't believe in miracles, like an implementation without
autotools. it's a bit sad that you want to get rid of it before we have
a replacement -- no matter if it exists.

similarly: i sometimes use gcc to compile stuff. i often use processors
to compute something. there's no way to test everything, even if such
practice might reveal some bugs.

cheers
felix