Two more AI bugs

[Hans Ronne]

>> Perhaps one of the existing test games could be adapted for this?
>
>Hmm, I took another look and those tests would seem to be closer than
>I realized to being automated.  At least some of them.  I checked in a
>few makefile and cvsignore fixes to "make check".
>
>This found bugs.  One of them was that running
>
> ../kernel/skelconq -f ../test/all.g -L ../lib <libtest.inp
>
>from the test directory causes a core dump.  I didn't diagnose it
>further, but I'd guess that the code is confused about how big the
>world is.  The all.g file looks like the following which gives one
>width and then later changes it:
>
>(area 20 20
>  . . .
>  (width 30)
>  . . .
>  (height 25)
>  . . .
>)
>
>This probably can be an error, but it shouldn't be a core dump.

I know. Fixing this crash is actually the first item in the TODO list that
Stan left for the next release. However, running all.g causes a hard freeze
under MacOS so I haven't got around to debugging it yet. If you can figure
out what the problem is, it would be great.

Hans

Hans Ronne

hronne@pp.sbbs.se