Show HN: A Lisp Interpreter for Shell Scripting

5 days ago (github.com)

Redstart is a lightweight Lisp interpreter written in C++ with a focus on shell scripting. It lets you combine the expressive power of Lisp with the practicality of the Unix shell: you can run commands, capture output, pipe between processes, and still use Lisp syntax for logic and structure. Think of it as writing your shell scripts in Lisp instead of Bash.

23 comments

quintussss

Reply
vindarel  1 day ago

> writing your shell scripts in Lisp instead of Bash.

in a Lisp ;)

Related, for Common Lisp:

- unix in lisp https://notabug.org/quasus/lserver/ - live-coding remote function calls for the shell. Write a command in the REPL, and run it instantly in the shell.

I use CIEL ;)

And, built-in: use the --load flag or build a self-contained binary, compiled to machine code with SBCL. It can contain your web assets (html, js etc). A compressed binary weights ±30MB and starts fast. A stripped off binary with LispWork$ (no compiler, no debugger etc) is ±5MB. There's ECL too.

esrh  1 day ago

awesome! I have wanted something like this for a long time. Currently I use a janet fork <https://github.com/eshrh/matsurika> with some trivial additions, the most important of which is a `$` macro that does what the `sh` does here. I have two questions:

- I see that `sh` does not take in strings but instead lisp forms. How do you distinguish between variables that need to be substituted and commands? In my fork, the way to do variable substitution involves quasiquoting/unquoting. - Almost all of the features that make your language good for shell scripting are essentially syntactic features that can easily be implemented as a macro library for say, scheme. Why'd you choose to write in C++? Surely performance is not an important factor here. (I'm interested because I am currently working on a scheme-based shell scripting language).

  • em-bee  1 day ago

    can you give an example of how variable substitution in your language looks like?

    one of the things i think a lisp for shell should have, and i agree that this may not be easy, but unix commands should be first class functions, as in, you should not need a $ or sh macro to make them work. the other thing is that strings should not be quoted, and so you need something else to designate variables like $path or ($ path)

    • esrh  19 hours ago

      Yes, i agree that unix commands should be first class. I did this for the super common stuff like ls and cp. As for substitution, I did exactly $ for substitution. You'd do something like ($ rsync -avP $src $dst), but I don't think I ever got around to implementing $() to evaluate forms. If you really need to do that then you have to quasiquote the whole expression and unquote the form you need to evaluate. This has been relatively ok for me though. I never implemented anything like pipes or redirection, I instead just send everything like that to bash.

      This is not really relevant to your question, but I regret choosing janet for this, it's too opinionated and hacking on C is not as fun as lisp. I started writing my own version of schemesh in racket, but I never got far enough.

em-bee  1 day ago

this made me wonder why guile isn't designed for shell scripting too. i mean, wouldn't that make sense? if you want guile to be the designated extension language for gnu applications, why not also make it the designated language for shell scripts?

ctenb  1 day ago

Very nice! I've often wondered how close you could get to a POSIX-like syntax with something like this while maintaining a LISP semantics as much as possible. Especially pipelines are much easier to read with the | and > operators. I guess you need some sort of LISP dialect that supports infix operators

  • sshine  1 day ago

    You can have regular shell infix pipes combined with Lisp/Scheme macros as control flow. I think the tradeoff that Schemesh is nice, even though it does sacrifice POSIX:

    The best of both worlds of shell and Lisp is quick ability to run and pipe processes, and full programming functionality without the shell scripting shenanigans like obscure semantics and lack of good data structures.

  • cess11  1 day ago

    Unless someone has already provided a library for it, write some wrappers around pipes and forks and use OCaml utop.

    It's not a Lisp but close enough, I'd say. If I didn't have the rather extensive background of using Picolisp and some other REPL-like tools as a form of shell I'd probably have settled for utop, at least until I reached my iex era.

antics9  1 day ago

This looks good! I've seen other tries of shell scripting with lisp dialects but Redstart syntax looks more intuitive (from a shell scripting standpoint) and easy to read.

waynesonfire  19 hours ago

This misses the point of shell scripts.

OUTPUT=$(some_command)

is the killer app of sh scripts among other things, like being universally available. Any language, including this, can execute a command. It's the ergonomics that matter. Shell scripts reduce the impedence of interacting with the shell to control the OS. At first sight, this tool does'nt improve on that impedendance. It also invents it's own esoteric syntax, so what's the gain?

On a separate point but equally important, I don't understand why sh scripts are exempt fom unit tests. Do these tools alleviate that?

  • suspended_state  13 hours ago

    > On a separate point but equally important, I don't understand why sh scripts are exempt fom unit tests.

    Likely because:

    - it's an older culture than unit testing

    - people write shell scripts for one shot tasks, or to automate tasks which are system setup/configuration specific

    - if a shell script becomes useful enough, it gets rewritten in a more apt language