On Thu, Aug 11, 2011 at 12:38:17PM -0800, Roger wrote:
> OK. Now I see the collapsing, and it seems more like a hidden collapse rather
> then an immediately apparent collapse from an initial stance after
> reading/tracing. When tracing this function, I was thinking the function
> would be read each time the top function was called.
I'd call it "replacing the function with a different one". And yes,
at first glance it looks like the function is defining a new function
(with the same name) recursively, in preparation for calling it; but
apparently not. It's wiping itself out so the *next* call gets the
new function.