Re: [chessx-developer] Big commit into svn - new Board() implementation

[Sean <sea...@sy...>]

On Mon, 28 May 2007 20:24:46 +0200
"Bruno Rizzuti" <bru...@gm...> wrote:

> Such an error doesn't appear for me (WinXP+Qt4.2.2+rev.644), instead
> it compiles and runs with no problem at all. On the other hand, tens
> of warnings appear during the compilation, all with basically the same
> message. I am attaching a screenshot with a sample, you might want to
> eliminate them.

Definitely want to eliminate such warnings.  Will have to install a cross compiler
so I can figure out what it's complaining about.  Will follow up on this shortly.

> This does not compile for me. Just to be clear, what I did in this
> case is "qmake CONFIG+=fastbits" followed by "make release", instead
> of "qmake" followed by "make release". I am attaching a second
> screenshot with all the error messages I get, they probably will give
> you a hint of what is happening.

You did everything correctly and it's not surprised it didn't compile.  If it's
possible for you to try the attached patch, to see if it lets you compile with
the fastbits option that'd be interesting and helpful.

> I didn't test the SetupPosition yet. I will, but I also hope that in
> the future it would be possible to fulfill my feature request n.
> 1715720 (Setup Position by drag-and-drop), which may make the
> SetupPosition operation more intuitive IMHO.

The way it is done now actually lets you set up a board much faster than
by dragging and dropping.  But eventually we should be able to allow
setup to be done either way, users choice.

> About the bug n.1716388 (Highlighting active for immobilized pieces),
> which BTW you didn't close yet... well, of course what was left wasn't
> a bug anymore, but rather a feature request. In this respect, I would
> suggest to improve it further. For instance, in the following position
> "rnbqk2r/pppp1ppp/5n2/4p3/4P3/BP3N2/P1PP1PPP/RN1QKB1R b KQkq - 2 4 "
> it would be nice if the black king would be considered immobilized as
> well.

It's closed now after adding a bug-fix into SVN for this problem you spotted;
thanks very much.

Sean.