Hello!

The attached patch is a first stab at re-implementing SRFI-1 in Scheme.

Here’s a quick benchmark of ‘fold’, for large and small lists:

  - in C

    ("srfi-1.bm: fold: fold" 30 user 5.55 benchmark 5.54999599456787 bench/interp 5.54999599456787 gc 0.0)
    ("srfi-1.bm: fold: fold" 2000000 user 4.41 benchmark 4.14297119140625 bench/interp 4.14297119140625 gc 0.0)

  - in Scheme (debug engine)

    ("srfi-1.bm: fold: fold" 30 user 6.04 benchmark 6.03999599456787 bench/interp 6.03999599456787 gc 0.0)
    ("srfi-1.bm: fold: fold" 2000000 user 5.14 benchmark 4.87297119140625 bench/interp 4.87297119140625 gc 0.0)

  - in Scheme (regular engine)

    ("srfi-1.bm: fold: fold" 30 user 5.46 benchmark 5.45999656677246 bench/interp 5.45999656677246 gc 0.0)
    ("srfi-1.bm: fold: fold" 2000000 user 4.64 benchmark 4.4111181640625 bench/interp 4.4111181640625 gc 0.0)

IOW, with the debug engine (currently the default) and for large lists
‘fold’ in Scheme is ~9% slower than in C; for small lists it’s ~17%
slower.

With the regular engine, Scheme is ~2% faster for large lists and still
~5% slower for small lists.

I’m tempted to put this in and then make the regular engine the default
unless ‘--debug’ is specified.

What do you think?

Thanks,
Ludo’.