LANGUAGE_SUPPORT.md

Language Support

Supported Languages

Language	Extensions	Parser	Symbol Types	Decorators	Docstrings	Notes / Limitations
Python	.py	tree-sitter-python	function, class, method, constant, type	@decorator	Triple-quoted strings	Type aliases require Python 3.12+ syntax for full fidelity
JavaScript	.js, .jsx	tree-sitter-javascript	function, class, method, constant	—	// and /** */ comments	Anonymous arrow functions without assigned names are not indexed
TypeScript	.ts, .tsx	tree-sitter-typescript	function, class, method, constant, type	@decorator	// and /** */ comments	Decorator extraction depends on Stage-3 decorator syntax
Go	.go	tree-sitter-go	function, method, type, constant	—	// comments	No class hierarchy (language limitation)
Rust	.rs	tree-sitter-rust	function, type (struct/enum/trait), impl, constant	#[attr]	/// and //! comments	Macro-generated symbols are not visible to the parser
Java	.java	tree-sitter-java	method, class, type (interface/enum), constant	@Annotation	/** */ Javadoc	Deep inner-class nesting may be flattened
PHP	.php	tree-sitter-php	function, class, method, type (interface/trait/enum), constant	#[Attribute]	/** */ PHPDoc	PHP 8+ attributes supported; language-file <?php tag required
Dart	.dart	tree-sitter-dart	function, class (class/mixin/extension), method, type (enum/typedef)	@annotation	/// doc comments	Constructors and top-level constants are not indexed
C#	.cs	tree-sitter-csharp	class (class/record), method (method/constructor), type (interface/enum/struct/delegate)	[Attribute]	/// <summary> XML doc comments	Properties and const fields not indexed

---

Parser Engine

All language parsing is powered by tree-sitter via the tree-sitter-language-pack Python package, providing:

• Incremental, error-tolerant parsing
• Uniform AST representation across languages
• Pre-compiled grammars for supported languages

Dependency: tree-sitter-language-pack>=0.7.0 (pinned in pyproject.toml)

---

Adding a New Language

1. Define a LanguageSpec in src/jcodemunch_mcp/parser/languages.py:

NEW_LANG_SPEC = LanguageSpec(
    ts_language="new_language",
    symbol_node_types={
        "function_definition": "function",
        "class_definition": "class",
    },
    name_fields={
        "function_definition": "name",
        "class_definition": "name",
    },
    param_fields={
        "function_definition": "parameters",
    },
    return_type_fields={},
    docstring_strategy="preceding_comment",
    decorator_node_type=None,
    container_node_types=["class_definition"],
    constant_patterns=[],
    type_patterns=[],
)

2. Register the language:

LANGUAGE_REGISTRY["new_language"] = NEW_LANG_SPEC

3. Map file extensions:

LANGUAGE_EXTENSIONS[".ext"] = "new_language"

4. Verify parser availability:

from tree_sitter_language_pack import get_parser
get_parser("new_language")  # Must not raise

5. Add parser tests:

def test_parse_new_language():
    source = "..."
    symbols = parse_file(source, "test.ext", "new_language")
    assert len(symbols) >= 2

---

Inspecting AST Node Types

To inspect the node types produced by tree-sitter for a source file:

from tree_sitter_language_pack import get_parser

parser = get_parser("python")
tree = parser.parse(b"def foo(): pass")

def print_tree(node, indent=0):
    print(" " * indent + f"{node.type} [{node.start_point}-{node.end_point}]")
    for child in node.children:
        print_tree(child, indent + 2)

print_tree(tree.root_node)

This inspection process helps identify the correct symbol_node_types, name_fields, and extraction rules when adding support for a new language.