tree-sitter-chicory

This repository contains a Tree-sitter grammar for the Chicory programming language. Tree-sitter is an incremental parsing system that can build a syntax tree for source files and efficiently update the syntax tree as the source file is edited.

Installation

Helix

1. Add this to your Helix languages.toml:

[[language]]
name="chicory"
scope="source.chic"
file-types=["chic"]

[[grammar]]
name = "chicory"
source = {git = "https://github.com/chicory-lang/tree-sitter-chicory" , rev="CHANGE ME: target commit hash, probably latest for now"}

2. Run hx -g fetch to update the grammar
3. Run hx -g build to build the grammar
4. Copy the queries into ~/.config/helix/runtime/queries/chicory

Neovim

1. Add chicory to the list of ensure_installed languages in your require('nvim-treesitter.configs').setup block.

2. Associate .chic files with chicory:

vim.filetype.add({
  extension = {
    chic = "chicory", -- Map `.chic` files to the `chicory` filetype
  },
})

3. Tell neovim where to find the Chicory parser:

local parser_config = require("nvim-treesitter.parsers").get_parser_configs()
parser_config.chicory = {
  install_info = {
    url = "https://github.com/chicory-lang/tree-sitter-chicory.git", -- Replace with your grammar repository
    files = { "src/parser.c" },                                      -- Add other files like `src/scanner.c` if needed
    branch = "main",                                                 -- Replace with the appropriate branch
    generate_requires_npm = false,                                   -- Set to true if your grammar requires npm
    requires_generate_from_grammar = false,                          -- Set to true if you need to generate the parser
  },
  filetype = "chicory",                                              -- Associate the parser with your language's filetype
}

4. Put/link the queries in the right place (I used a symlink):

$ ln -s .config/nvim/queries/chicory $TREE_SITTER_CHICORY/queries

Note: You can also point the parser to a local copy of the tree-sitter-chicory repo instead of the github url.

For more information, see https://github.com/nvim-treesitter/nvim-treesitter/#adding-parsers

ANTLR Grammar Implementation Progress

This section tracks the implementation of major language features from the ANTLR grammar.

Implemented

Types

• Primitive types (number, string, boolean)
• Tuple types (e.g., [number, string])
• Record types (e.g., { x: number, y: number })
• Algebraic Data Types (ADTs)
  • Simple variants (e.g., type Shape = Circle | Square)
  • Variants with primitive types (e.g., type Option = None | Some(number))
  • Variants with record fields (e.g., type Result = Ok(string) | Error({ message: string }))
• Type references (uppercase identifiers)

Expressions

• Match expressions (e.g., match (value) { Some(x) => x + 1, None => 0 })
  • Pattern matching on variants
  • Pattern matching with literals
  • Pattern matching with identifiers
• Binary operations
• Function calls
• Literals (number, string, boolean)
• Identifiers
• Parenthesized expressions
• JSX expressions
  • Self-closing elements (e.g., <Div />)
  • Elements with children (e.g., <Div>Hello world</Div>)
  • Elements with attributes (e.g., <Div className="container" count={5} />)
  • Nested elements
  • Expression children (e.g., <Div>{count}</Div>)

Statements

• Let assignments (e.g., let x = 42)
• Const declarations (e.g., const PI = 3.14)
• Assignment statements
• Import statements
• Export statements

Other Features

• Comments (single-line and multi-line)