Re: [AD] Problem with remove_mouse() and video bitmaps.

[Stef <stapostol@xxxxxxxxxx>]

> > True, but shouldn't the screen always be valid (while there is a gfx  
> > mode
> > set)? I think that the problem was with video bitmaps, which the above
> > code would cover.
>
> Yes, "screen" was just an example for any bitmap (the mouse might be
> drawn to). What we are saying is, you simply can't set the bitmap to
> NULL, without some extra construction to keep track of all active bitmap
> pointers.

I don't see why this should be a problem. Especially since the  
documentation doesn't state that destroy_bitmap() won't change the pointer  
it's given.

That the solution is not robust (and can't be robust without any extra  
construction) is a given, but I'd say it is better than the current  
situation (it would at least stop some random crashes).

@Peter:
> Set _which_ pointer to NULL?

void destroy_bitmap(BITMAP *bitmap)
{
   /* ... */
   bitmap = NULL;
}

That one.