Re: [Gnushogi-devel] Comparing GNUShogi 1.2 and later versions

[H.G. Muller]

Op 2/18/2018 om 4:04 PM schreef address@hidden:
> For this my Omaha UI could be of use, as it's able to talk both protocols,
> but i've not worked on it for some time now, and it's still unable to
> handle clocks, which makes it of limited use at best.  Maybe I should have
> a look at this, it was what I last worked on on this software, I just
> remember it was not finished...
>
> The obvious other idea would be to backport your XBoard patches, but then
> this work would be of limited use aside from checking the AI sanity,
> so I'm less inclined to go this path :}
Hmm, neither option sounds very attractive. (Unless the inteface code 
was changed so
little between v1.2 and v2.3 that the patch applies without protest.) 
Making the UI
manage clocks is of course worth it, but it will be a lot of work.

I don't exactly remember how XBoard and xShogi protocol differed. 
Obviously in move
notation and swapping of the black and white commands, and otim -> 
otime. And
probably the setting of time control might be different. Perhaps it is 
possible to adapt
UCI2WB to become an XS2WB adapter with very little work. It already 
parses all CECP
commands though a sequece of string compares, and I can just delete the 
handling of
 all commands that don't need modification, and pass unrecognized commands
unchanged to the engine. Routines to convert move notation are already 
in there (for USI).
The routine Engine2GUI would have to be completely rewritten to parse 
xShogi protocol
engine->GUI commands, but there are only very few of those. (Basically 
just moves
and thinking output, and we don't care about the thinking output.)

Of course this would still be a 'dead-end project', but it might be (by 
far) the simplest
of all options. And it would work for all GNU Shogi versions.
> Great news, I guess I can find a slot for such a small task :)
> Thanks!
As to CrazyWa: I hope you haven't spent any time on this yet, as I just 
discovered that
Tori Shogi was broken (already in the version I started the cleanup 
from). At some point
I insertd an alias variant name for washogi (11x17+16_chu), but I had 
not shifted the
corresponding data in tables indexed by the variant number, so that the 
Wa alias was
using the Tori values, and Tori Shogi was using null data. I discovered 
this when I tried
to add yet another variant (Euro-Shogi), which CrazyWa already claims it 
can play.

I pushed the Tori/Wa fix to my on-line repository, but perhaps I should 
add euro-Shogi
tables as well, so that it indeed plays this. So give me oe more week to 
see if this is as
easy as I think it should be.
 Regads,
H.G.