Bug#688934: Another way to trigger this bug

[Sylvain]

Hi Bas,

On Sun, Oct 14, 2012 at 12:40:37PM +0200, Bas Wijnen wrote:
> On Sun, Oct 14, 2012 at 11:49:43AM +0200, Sylvain wrote:
> > At the very least, let's tag the miswrites in the code with a TODO.
> > If you tell me you're perfectly sure this has no side effect, we can
> > fix them as well, but I'd suggest doing that only if we're sure.
> > Comments?
> 
> I'm not perfectly sure for all other things, but the animation part
> cannot have been used, because it would trigger this hang (which was
> also reported from original Dink).

Indeed.

> Also, I can't really imagine how a
> dmod would ever want to make use of the fact that spr[0] gets written
> to. The dmod can't even use these values: if you ask for sp_seq (0, -1)
> it will return -1, no matter what is in there.

Indeed at first glance all sp_XXX() functions are protected against
using spr[0].

This leaves indirect bugs which, instead of freezing the game like
this one, would just produce a side effect.

> Still I understand your reasoning, and would be fine with this fix as
> well.

Plus, apparently there're several other unprotected calls to
'add_sprite' :/

I tagged the calls you pointed out, and I'll leave the code as-is.

-- 
Sylvain