gnugo-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gnugo-devel] Re: cgoban1 maintainership


From: Gunnar Farneback
Subject: Re: [gnugo-devel] Re: cgoban1 maintainership
Date: Mon, 28 Oct 2002 17:46:59 +0100
User-agent: EMH/1.14.1 SEMI/1.14.3 (Ushinoya) FLIM/1.14.2 (Yagi-Nishiguchi) APEL/10.3 Emacs/20.7 (sparc-sun-solaris2.7) (with unibyte mode)

Kevin wrote:
> > (2) It would be good if GTP (Go Text Protocol) support
> > could be hacked into CGoban1. You can probably get
> > help for this from the GNU Go and GTP lists.
> 
> GTP is probably a feature request, and we'll look into that for a later
> release, since I want to get some existing bugs & packaging issues out
> of the way on 1.9.x before adding features, y'know? 

It would definitely be a new feature. From cgoban's point of view it's
an alternative to GMP and probably not hard at all to implement (for
someone who's familiar with the code, that is). Compared to GMP it
also makes it practical to implement continuation of an earlier saved
game, which would probably be a popular feature.

Another new feature I would like to add to the wish list is support
for viewing diffs of sgf files. GNU Go stores its joseki database in a
couple of largish sgf files. Unfortunately the sgf format makes it
practically impossible to review a raw diff of modifications. One
alternative is to read both original file and patched file into an sgf
viewer, but unless the file is small it's still almost hopeless to
locate the modifications. (*)

What I'm dreaming about is a cgoban sgf tree view where the
differing nodes are marked by a change in background color. It would
also be helpful to change the color of the edges in the graph leading
to difference areas.

This is probably not at all easy to implement but it would be
extremely useful for the maintenance and development of the joseki
database in GNU Go.

/Gunnar

(*) The most feasible approach today is probably to run both original
and patched files through the joseki compiler to get them into the .db
format and then make a standard diff. A problem is that we usually get
a big false delta from patterns which have only changed number. 




reply via email to

[Prev in Thread] Current Thread [Next in Thread]