[plt-scheme] What's the latest on the Scsh port?

[I'm in the middle of moving, no home internet yet etc, so this is
just a limited quick reply...]

On Jul 12, Neil W. Van Dyke wrote:
> One very simple thing would be for the command line parser to read
> the first regexp in the line, and if it's a symbol bound to a
> procedure, than evaluate the input as if it were surrounded by
> parens.  (Alternatively, the command prompt could initialize the
> input to "()" and leave the point in the middle, I suppose.)  [...]

There's another important bit -- if you read an "ls" sexp, then you
need to do the right thing even if it is unbound.  I think that this
issomething useful in general so unless you bind it, using `ls' is
possible as a scheme function -- `map' would be way better-looking
than `xargs'.  Generally, when I think about stuff I need my shell to
do, I think that it is most likely easier to have some alternative
(more shell-like) grammar for interactions with easy escaping back to
scheme, and the whole thing is completely redundant for scripts.


> I'd try to use one of the portable and solid but minimal Unix shells
> as a starting point to get the esoterica of TTY ownership, process
> groups, etc. on different Unices right.  [...]

I plan to avoid any C-level hacks, especially ones that will be system
dependent.  The idea should be that you have some stable+convenient
Scheme environment to work with, and if you want some of theabove,
you'll need to add some extension (which should be trivial using a
good ffi thing...)


> I would also keep in mind that at some point this interactive
> command shell might be incorporated into an mzwm... :)

That's another idea that I (obviously) played with...  But looking at
sawfish (and at scwm before that) I get the feeling that with the
whole low-level X mess, with the added headaches of numerous
specifications (that many-C thing, Gnome, etc), it's just not worth
it.  I just get a .sawfishrc that works fine for me, and every time I
get into some artifact of the Emacs-mentality, I try to think happy
thoughts in an attempt make the hacking urge go away.


> MzScheme is already good for "scripting", as far as I've noticed.
> For Unix, it just could use a few more Posix functions.  [...]

What I have in mind is mostly Scheme stuff -- like easily piping stuff
together (both subprocesses and Scheme threads), and a whole bunch of
small things I keep around for the many scripts I have.


> I did suggest a simple regexp-binding macro in the PLT bug report
> system; [...]

Yeah -- things like this.  (BTW,
http://www.cs.cornell.edu/eli/Swindle/misc-doc.html#regexper)


> (Although I still think that the conventional shell scripting
> paradigm of chaining shell programs together with lots of haphazard
> parsing glue should die a quick but painful death, which would
> eliminate a lot of the need for regexps.)

Should, but in the meantime, I want it to beconvenient.  (I expect
this "meantime" will probably be very long...)


> I still really like how Python .py/.pyc and Emacs .el/.elc file pairs
> work, but I guess MzScheme's "compiled" subdir and "setup-plt -l" are
> close to as convenient for scripts.

What's the point?  Most programs that can be called "scripts" should
work just fine as a source file -- the above holds for libraries,
which is part of mzscheme already...


> The only two hesitations I've had about using MzScheme for scripting
> are: (1) I currently only have MzScheme from CVS installed, and I've
> broken it for a day 2 or 3 times;

Heh, I learned that also -- I keep a default stable plt tree for
scripts.


> (2) every machine I use would need to have PLT installed for my
> scripts to work.  Both of these issues are mostly resolved for me,
> now that a recent version of MzScheme is in Debian.

It should be possible to get an executable with the embedded code in
so you just have one file to copy, no?

-- 
          ((lambda (x) (x x)) (lambda (x) (x x)))          Eli Barzilay:
                  http://www.barzilay.org/                 Maze is Life!