Re: [gnugo-devel] fast_defense() improvement

[Paul Pogonyshev]

nando wrote:
> For Paul: "proving" the safety looks difficult. It would involve reading,
> but that's precisely what we would like to avoid.

You misunderstood me.  I meant a formal proof of safety of the proposed
code.  Or at least an informal one, with diagrams and all.  I'd like to
see some diagrams explaining how it works and why and in which circumstances
we need to subtract liberties.

You can also confirm safety non-directly by say disabling fast_defense(),
but comparing its results with actual reading results.  And then running
this through regressions.  Along this lines:

        if (fast_defense(...))
          assert(find_defense(...) == WIN);

Paul