command line shells
Software Scavenger
I.e. you would be able to interact with it like a normal shell such as
the Korn Shell, giving it commands such as "du -sk | sort -n", but you
could also enter top-level Lisp forms, and it would act like a REPL.
I would prefer one with versions for Unix, Linux, and Windows, so I
could use the same shell on each.
If there aren't any such shells available, is it because Common Lisp
users tend to not need shells much, or because it would be hard to
implement and nobody has had time for it yet?
Dr. Edmund Weitz
This isn't quite what you're looking for but a nice hack nevertheless:
<http://clisp.sourceforge.net/clash.html>.
Edi.
Alberto Riva
<http://www.swiss.ai.mit.edu/ftpdir/scsh/scsh.html>
or
<http://scsh.sourceforge.net/>
Erik Naggum
> Are there any good command line shells available based on Common Lisp?
* Alberto Riva
| Try scsh, the Scheme shell.
I was not aware that the Scheme shell had acquired a Common Lisp "mode".
Perhaps you are not aware that Common Lisp is not a dialect of Scheme?
The tendency among some people to answer with Scheme solutions when the
questions _specifically_ requires a Common Lisp answer is very strange
and very, very annoying. Recently, I met a person who thought this was
OK because he could not fathom that "Lisp" was more than Scheme, and he
refused to accept facts to the contrary because his professor had told
him that Scheme is Lisp. This must be intentional by the Scheme freaks.
///
--
The past is not more important than the future, despite what your culture
has taught you. Your future observations, conclusions, and beliefs are
more important to you than those in your past ever will be. The world is
changing so fast the balance between the past and the future has shifted.
Thomas F. Burdick
> * Software Scavenger
> > Are there any good command line shells available based on Common Lisp?
>
> * Alberto Riva
> | Try scsh, the Scheme shell.
>
> I was not aware that the Scheme shell had acquired a Common Lisp "mode".
>
> Perhaps you are not aware that Common Lisp is not a dialect of Scheme?
Heh. While Common Lisp is not a dialect of Emacs Lisp, either, the OP
may also be interested in eshell, which comes with recent Emacs
releases. Unlike Scheme, you can type quite a lot of things into a CL
repl or an elisp repl and get the same result. There are even more
things that you can type very similarly to get the same result.
Especially if you (require 'cl). At an eshell prompt, you can type
`ls' or:
(loop for i from 0 to 9
do (sleep-for 1)
collect (format "At time T%d the date is %S"
i (eshell-command-result "date")))
and they'll both do what you expect.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
Alberto Riva
>
> * Alberto Riva
> | Try scsh, the Scheme shell.
>
> I was not aware that the Scheme shell had acquired a Common Lisp "mode".
> Perhaps you are not aware that Common Lisp is not a dialect of Scheme?
I am well aware of that; I just read the original question too quickly.
I also assumed that the OP was more interested in being able to use a
Lisp-like syntax than in the details of CL vs. Scheme. I'm sorry if I
caused any confusion. I am no "Scheme freak" and I use CL exclusively
for my work, but I still think that scsh is a very nice idea.
Alberto.
Kaz Kylheku
> and very, very annoying. Recently, I met a person who thought this was
> OK because he could not fathom that "Lisp" was more than Scheme, and he
> refused to accept facts to the contrary because his professor had told
> him that Scheme is Lisp. This must be intentional by the Scheme freaks.
Funny how some people have a one-shot trigger for accepting information.
Why is it that he didn't reject what his professor was saying? Is some
statement A more true than a statement B, simply because A was heard
before B?
After a typical undergraduate program in computer science, it takes
several years of active research and thinking to expunge the
bits of misinformation you have been fed among the good information.
Erik Naggum
| Funny how some people have a one-shot trigger for accepting information.
| Why is it that he didn't reject what his professor was saying? Is some
| statement A more true than a statement B, simply because A was heard
| before B?
It appears that my current .signature applies a lot more than I thought.
Indeed, many people believe the first statement they hear in some area,
because there is no cognitive dissonance when there is no information in
that area. When a later statement comes along that corrects a previous
statement, it does create cognitive dissonance, and many people have
never been taught how to deal with that feelilng, so they merely reject
that which causes it, instead of examining it. According to the theory
of cognitive dissonance, most people work hard to get rid of the feeling
and completely fail to understand how it works or why they experience it.
I have tended to believe that the ability to deal with this feeling in a
productive manner is a mark of intelligence.
| After a typical undergraduate program in computer science, it takes
| several years of active research and thinking to expunge the bits of
| misinformation you have been fed among the good information.
Unfortunately for me, I started the process of expunging the junk before
I had finished my studies. Working while studying to avoid student loans
has the side effect of giving you real life experience prematurely. That
can just _ruin_ an academic carreer.
Christopher Stacy
Software> Are there any good command line shells available based on Common Lisp?
Software> I.e. you would be able to interact with it like a normal shell such as
Software> the Korn Shell, giving it commands such as "du -sk | sort -n", but you
Software> could also enter top-level Lisp forms, and it would act like a REPL.
Software> I would prefer one with versions for Unix, Linux, and Windows, so I
Software> could use the same shell on each.
Software> If there aren't any such shells available, is it because Common Lisp
Software> users tend to not need shells much, or because it would be hard to
Software> implement and nobody has had time for it yet?
I think it would be fairly easy to do this: most implementations
provide some sort of streams-based interface which can be used to
run programs (either through another shell, or directly), they
also provide thread ("Lisp process") control. That just leaves
input editing, which is also generally already available as part
of the implementation's GUI-building features.
The Lisp Machine operating system, of course, had a shell that accepted
both commands (with very fancy command completion) or Lisp forms.
James A. Crippen
> In article <32168602...@naggum.net>, Erik Naggum wrote:
> > and very, very annoying. Recently, I met a person who thought this was
> > OK because he could not fathom that "Lisp" was more than Scheme, and he
> > refused to accept facts to the contrary because his professor had told
> > him that Scheme is Lisp. This must be intentional by the Scheme freaks.
Hmm, perhaps, though I have a hard time understanding such
inflammatory invective.
Scheme is *a* Lisp. Common Lisp is also *a* Lisp. Lisp 1.5 is also
*a* Lisp. Lisp is a family of languages. Scheme certainly isn't
*the* Lisp, but I doubt that any Scheme programmer claim that. There
are however folks who have claimed that Common Lisp is *the* Lisp.
Such opinions have not been particularly well received, I think,
particularly by those who have used other Lisps.
'james
--
James A. Crippen <ja...@unlambda.com> ,-./-. Anchorage, Alaska,
Lambda Unlimited: Recursion 'R' Us | |/ | USA, 61.20939N, -149.767W
Y = \f.(\x.f(xx)) (\x.f(xx)) | |\ | Earth, Sol System,
Y(F) = F(Y(F)) \_,-_/ Milky Way.
Kent M Pitman
> k...@ashi.footprints.net (Kaz Kylheku) writes:
>
> > In article <32168602...@naggum.net>, Erik Naggum wrote:
> > > and very, very annoying. Recently, I met a person who thought this was
> > > OK because he could not fathom that "Lisp" was more than Scheme, and he
> > > refused to accept facts to the contrary because his professor had told
> > > him that Scheme is Lisp. This must be intentional by the Scheme freaks.
>
> Hmm, perhaps, though I have a hard time understanding such
> inflammatory invective.
I don't personally find this inflammatory. Let me say why: I think it
is just as divisive to take your position as Erik's. Pitman's Rule of
Political Interpretation: "There are no political answers, only
political questions". If Erik's answer to the quesiton is political,
so is yours. And in all likelihood, for every group A who thinks
group B's solution to some political question Q is inflammatory, group
B thinks (or, at least, is entitled to think) that group A's answer to
the same question Q is inflammatory. So either both his position and
yours are inflammatory, or neither are. I prefer neither, but, as
they say, YMMV.
> Scheme is *a* Lisp. Common Lisp is also *a* Lisp. Lisp 1.5 is also
> *a* Lisp. Lisp is a family of languages. Scheme certainly isn't
> *the* Lisp, but I doubt that any Scheme programmer claim that. There
> are however folks who have claimed that Common Lisp is *the* Lisp.
> Such opinions have not been particularly well received, I think,
> particularly by those who have used other Lisps.
These are opinions.
I personally think it's possible for someone to take the position that
Scheme is a Lisp, but I think it's its own unique language family,
enough separate that it should not be called a Lisp. This is not a
slight in any way. It is a statement of my personal perception of
relative conceptual difference. The overlap between Scheme and most
other Lisps is so small, and the gap so large that I would compare it
to the C++ divide. (Not all aspects of that analogy work. But the
sense of distance does. It is not fair to call Java a C or a C++...)
So call Scheme a lisp if you like, but don't assume that everyone must
nor does. Certainly the fact that CL is not "*the* Lisp" doesn't mean
you have, by saying so, supported the claim that Scheme is a Lisp
(even if such a claim is supportable by some other means).