Re: @bind

[Neil Jerram]

>>>>> "Marius" == Marius Vollmer <address@hidden> writes:

    Marius> Neil Jerram <address@hidden> writes:

    >> I could have a go, but it is the evaluator ... :-(

    Marius> Do fear it not, as it is our fear that makes it strong. ;)

But master, surely we want it to be strong?

    >> This is IMO much cleaner than how @bind works now.  Do you
    >> think you'll have time for this soon?

    Marius> I will get started now. [...and later...] Done.

Thanks for this.  I've been playing around with elisp and thought
others might be interested in taking a look or commenting on the
approach - hence the attached tarball.

I've started from Mikael's transformer, with a few additions and
bugfixes, and then see what happens if we switch into elisp syntax and
do `(load loadup.el)' - which is what the `bare' Emacs does before
dumping.  What happens is that you get so far before you find a
transformer bug - in which case fix the bug - or an undefined
primitive - in which case write a Scheme implementation of that
primitive (this is proof of concept - we can reimplement in C later if
necessary).  So far, byte-run.el and subr.el are OK, and we get stuck
on `string-match' in version.el.

If you have any comments, I'd be happy to hear them.

        Neil

elisp.tar.gz
Description: elisp files