George Ogata <g_ogata / optushome.com.au> wrote:

> I think Avi's doesn't quite meet the requirements.  This should be true, 
> shouldn't it?:
> 
> one_in_each([1,2,3],[[1,2],[1,2],[1,3]])

Ah, nice catch.  The only way I can think to fix that would be to use
reject{|e|e.id==s.id} as you do, instead of rows-[s].  Along with
Pit's suggestion, and adding a != nil, that would make our solutions
identical - so I think I'll just leave mine in its (broken) state.