Closh runs ClojureSript on node.js via lumo REPL. In order to be somewhat bashward compatible there is a command mode which transforms top-level forms in a macro-like way.
Thanks to Clojure's syntax for symbols supporting almost all characters we don't need to roll out a custom parser to support convenient unquoted notation for filenames and switches. The only customization done to the built-in reader is the support for multiple slashes in a symbol, which is required for nested directories.
To get an idea of the scope of an implementation, here is a basic overview of closh components:
- reader - Since most shell operators can be just read as symbols, writing a custom reader can be avoided so reading is done with
tools.reader. It needed minor tweaks to read things like IP addresses (e.g.8.8.8.8which would be an invalid number) or paths (e.g./a/path/to/filewhich would be an invalid symbol). Thanks totools.readerbeing written in Clojure the customizations were fairly simple. - parser -
clojure.specis used for parsing, specs look like a grammar for parser generators. Doing conform transforms the input into AST. - compiler - This is just a data transformation using plain clojure functions. Generated code uses the
pipeline. - pipeline - Functionality to run sequence of processes and pipe data.
- platform - Abstracted functions to spawn processes and redirect IO. Abstractions for JVM and node.js provide an almost identical API which the pipeline utilizes.
- frontend - This provides the repl for interactive use. For CLJS version we use internal node.js
readlinemodule. The CLJ version runs onclojure.main/replwithrebel-readline.
The reader customization, platform and frontends have separate implementations for CLJ and CLJS versions. The code for parser, compiler and pipeline is shared (using common CLJC files).