@codincod/codemirror-lang-raku

[ CHANGELOG ]

This package implements Raku language support for the CodeMirror code editor, using a Lezer grammar written for this package.

CodeMirror has never shipped Raku. CodeMirror 5 had a Perl 6 mode contributed in 2011 and dropped before version 6, and nothing replaced it, so a Raku file in a browser editor has been either plain text or Perl 5 with the wrong keywords.

Written in part for CodinCod, a competitive coding platform, where it colours the editor people solve puzzles in.

This code is released under an MIT license.

Usage

import {EditorView, basicSetup} from "codemirror"
import {raku} from "@codincod/codemirror-lang-raku"

const view = new EditorView({
  parent: document.body,
  doc: `say "Hello World!";`,
  extensions: [basicSetup, raku()]
})

Coverage

Statements, declarations, signatures and the expression grammar; classes, roles and grammars; phasers; the quoting constructs including heredocs; Pod; and the operator set with its meta-operators. Measured against 1614 Raku files from Rosetta Code, 83.83% parse with no error node.

Two things are deliberately left flat. The inside of an interpolated string is one token rather than a tree, and so is the body of a token, rule or regex: Raku's regex sublanguage is a language of its own and deserves its own grammar rather than a corner of this one.

Three things Raku decides for itself

The operators are not written out in the grammar. Raku's operator set is open, and its meta-operators build new ones out of old ones, so »+«, [+], Z+ and R- are operators that nobody listed anywhere. The tokenizer reads a run of operator characters, peels off the meta-operator wrappers and looks up what is left, then hands the parser one of seventeen precedence classes. A symbol it has never seen is an operator too, because anybody may write sub infix:<∧> and then use it.

Nothing is reserved. hyper introduces a statement and is also a method everybody calls; where is a trait and a perfectly good routine name. A word is a keyword only where the parser has room for that keyword, so the same word reads as a keyword in one column and as a name in the next.

A character means one thing where a value can follow and another where a value has just been read. That one question settles / (a pattern or a division), * (a whatever or a multiplication), < (a list of words or a comparison), x (a variable or a repetition), and the rest. The parser is asked which of the two it is in, and it answers by saying whether it has room for a multiplication yet.

Where even that is not enough, spacing decides, as it does in Raku itself: an infix has space on both sides or on neither, so space on the left alone means a term is starting. grep * %% 2, @a is a grep, and $exp +1 is an addition.

What it gives an editor

Syntax highlighting, folding for blocks, signatures and Pod, indentation that knows a brace closes a block and lines a signature up under its opening parenthesis, and completion for the keywords, the built-in types, the built-in routines and the dynamic variables.

rakuLanguage is exported for use with LanguageSupport, and parser for use on its own.

Testing it on your own code

node test/corpus.ts path/to/your/raku

It prints the files that failed, the first line of each that did, and the proportion that came out clean. The corpus itself is not included; Rosetta Code is CC BY-SA and cannot be redistributed under this licence.