Re: Removal of cell terrain (never a good idea!)

[mskala at ansuz dot sooke dot bc dot ca]

On Sun, 16 May 2004, Eric McDonald wrote:
> > > "terrain-damaged-type" with the obvious (t* t* 1) to give equal
> > > probability to all possibilities, the result is that it will sometimes

> > type for which there are no nonzero probability entries - problematic
> > because it's very easy to create that situation what with the table
> > defaulting to zero everywhere.  (For instance, if you say "yes" to terrain

> one. If you could provide some test cases with which to test your patch
> that would be good. I'm working on a different fix right now, but maybe

Attached is terrainbug.g, which demonstrates the first issue - just load
it up and hit "detonate", and you'll almost certainly get a fatal
error.  If you comment out line 30, you'll get a fatal error for a
different reason (the second issue I mentioned).  In both cases, the GDL
arguably is specifying something incorrect - but it seems like it would be
easy to make Xconq not die in such cases.
-- 
Matthew Skala
mskala@ansuz.sooke.bc.ca                    Embrace and defend.
http://ansuz.sooke.bc.ca/

Attachment: terrainbug.g
Description: Text document