Hi,
I've been meaning to implement a Scheme interpreter for quite some time now,
mostly for the fun of it (but also so I could get something like the BRL
running without needing the whole of Java - see <http://brl.sf.net/> and
<http://brl.sf.net/brl_4.html>).
I've since come across Neko and, impressed by its footprint and
performance, am
considering using it to build a Scheme compiler instead. I've been following
your project for a while now, and I have to say I like the direction it's
going. I do have a few questions, though.
First of all, let me say I welcome the license change. However, the GPL is
still mentioned in a couple of files, namely some of the standard library
source files like "libs/std/utf8.c". Was this by omission?
Second, from the point of view of a potential Scheme implementor, I
have to say
tail-calls were a much appreciated addition. I know call-cc has also been
mentioned, but I'd like to know how "committed" you are to implementing it
"someday" (no timelines required, just if it is in fact on the "todo list").
Then, and if I do decide to try and build this (I can make no promisses), how
would you welcome the contribution to Neko of "general purpose" libraries (that
I would undoubtedly have to write) for things like arbitrary precision
arithmetic?
And finally, one tiny thing I've come across: shouldn't the attribution
operator be right associative? I get this:
$ cat test.neko
a = b = 0;
$ nekoc test.neko
test.neko(1): Invalid access
While this works as (I) expected:
$ cat test.neko
a = (b = 0);
$ nekoc test.neko
Thanks in advance!
Nuno Cruces
--
Neko : One VM to run them all
(http://nekovm.org)
