Native, fast starting Clojure interpreter for scripting

babashka bash clojure graalvm scripting shell-scripting

Find a file

Michiel Borkent 97d3e34c91 add closh to the list of Clojure shell solutions		2019-08-09 23:30:39 +02:00
.circleci	initial commit	2019-08-09 15:43:27 +02:00
.clj-kondo	initial commit	2019-08-09 15:43:27 +02:00
doc	initial commit	2019-08-09 15:43:27 +02:00
resources	bump version	2019-08-09 23:20:47 +02:00
script	initial commit	2019-08-09 15:43:27 +02:00
src	add tests	2019-08-09 23:09:59 +02:00
test/babashka	add tests	2019-08-09 23:09:59 +02:00
.gitignore	initial commit	2019-08-09 15:43:27 +02:00
LICENSE	initial commit	2019-08-09 15:43:27 +02:00
project.clj	initial commit	2019-08-09 15:43:27 +02:00
README.md	add closh to the list of Clojure shell solutions	2019-08-09 23:30:39 +02:00
reflection.json	initial commit	2019-08-09 15:43:27 +02:00

README.md

babashka

A pure, fast and (severely!) limited version of Clojure in Clojure for shell scripting.

Properties:

pure (no side effects)
fast startup time
interprets only one form
reads from stdin and writes to stdout

Rationale

Most of your script is in bash, but you want a tiny sprinkle of Clojure.

If most of your shell script evolves into Clojure, you might want to turn to:

Status

Experimental. Breaking changes are expected to happen at this phase. Not all Clojure core functions are supported yet, but can be easily added. PRs welcome.

Installation

Linux and macOS binaries are provided via brew.

Install:

brew install borkdude/brew/babashka

Upgrade:

brew upgrade babashka

You may also download a binary from Github.

Usage

... | bb [--raw] '<Clojure form>'

There is one special variable, *in*, which is the input read from stdin. When the --raw flag is provided, *in* is a single string or vector of strings. When it is omitted, the input is read as EDN.

The current version can be printed with:

bb --version
0.0.3

Examples:

$ ls | bb --raw '*in*'
["LICENSE" "README.md" "bb" "doc" "pom.xml" "project.clj" "reflection.json" "resources" "script" "src" "target" "test"]

$ ls | bb --raw '(count *in*)'
11

$ echo '[1 1 1 1 2]' | bb '(vec (dedupe *in*))'
[1 2]

$ echo '[{:foo 1} {:bar 2}]' | bb '(filter :foo *in*)'
({:foo 1})

Functions are written using the reader tag #f. Currently up to three arguments are supported.

$ echo '3' | bb '(#f(+ %1 %2 %3) 1 2 *in*)'
6

Regexes are written using the reader tag #r.

$ ls | bb --raw '(filterv #f(re-find #r "reflection" %) *in*)'
["reflection.json"]

Find the line numbers where the word Clojure occurs using a case insensitive regex:

$ cat /tmp/test.txt
foo
Clojure is nice
bar
when you're nice to clojure

$ cat /tmp/test.txt | bb --raw '(map-indexed #f[%1 %2] *in*))' | \
bb '(keep #f(when (re-find #r"(?i)clojure" (second %)) (first %)) *in*)'
(1 3)

Test

Test on the JVM:

script/test

Although this tool doesn't offer any benefit when running on the JVM, it is convenient for development.

Test the native version:

BABASHKA_TEST_ENV=native script/test

Build

You will need leiningen and GraalVM.

script/compile

License

Distributed under the EPL License, same as Clojure. See LICENSE.