Bug#544874: fairymax does not work on big-endian machines

[h.g. muller]

Package: fairymax 4.8j-1

Fairy-Max unjustly claims stalemate on the first move, and refuses any 
input move when run on Mac Power PC.
The problem is a bug in the reading of the fmax.ini file, which reads char 
o[] and char oo[] with an int format.
This happens to work on little-endian machines, because the arrays are read 
low to high index,
but puts only 0 or -1 in them on a big-endian machine, which leads to an 
initial setup with no pieces,
and pieces that have no moves.

Upstream this is now fixed by declaring o[] and oo[] as int, rather than 
char (version 4.8K):
http://home.hccnet.nl/h.g.muller/dwnldpage.html

The same error is in the source file of the Xiangqi derivative maxqi.c.

H.G. Muller