frank at crystal-objects.com wrote:

> The problem reported by this bug/fix report from last August stiil
> exists in 3.7b, The fix proposed does prevent the inifinite loop. Is
> this the best way to go about this?
> 

I submitted a fix for this back in the fall, but have since changed my 
mind about the best thing to do in this situation.  The submitted fix, 
IIRC, added the collection parameter to a copy of self when it == self. 
  I'm now pretty convinced that an error should just be raised when the 
parameter == self (since it is almost definitely unintentional, and if 
you really wanted to do this for some reason you could always explicitly 
use a copy yourself).  Does this sound sensible?

Andrew