Re: self copying functions

[Graham Wakefield <lists@...>]

On Mar 15, 2007, at 4:29 PM, Miles Bader wrote:

> I thought that Graham wanted to copy it so he could set the fenv
> separately without disturbing the fenv of the original function -- but
> now I realize my method won't work either, as the wrapped function  
> will
> use its old fenv, not the wrapper's ...
>
> So I guess for his purpose, some sort of lower-level copy would be
> necessary.
>
> -miles

Exactly.

Upvalues should be duplicated along with the function.