Re: lexically scoped functions

[peter@robots.oxford.ac.uk]

> i'm pretty sure that this was not my reasoning when we decided that
> lexically scoped functions were not evaluated.  i don't know what the
> reasoning was;  it seems wrong to me now.  in particular, i want to
> point out that the efficiency issues are very minor.

I agree that it would be better to have the consistency as I said.  I was
just trying to come up to with justification for the current behaviour.  I
know that the efficiency is a minor issue, really my reasoning was based on
the fact that you know that you have redefined the function, so you can use
it via the variable reference.

> this behavior will not be changed in es-0.84; it does require some real
> work on the implementation.

When I was thinking of the efficiency, I suppose I was really thinking of
programmer efficiency and staying with the status quo.

> my tendency is to support lexically scoped functions in a later release
> but i want more reaction from the list before we go ahead and change.

I do think it's a good idea, and if you're keen to do it, that's even
better!

Pete.

P.S. When Noah said that es was the greatest ever (unix) shell, I
categorically agree with him.  Well done Paul and Byron for your hard work!