Function extract_pascal

graphify/extract.py:13348–13576

Extract units, classes, procedures, uses-imports, and calls from Pascal/Delphi files. Produces nodes for: - The file itself - unit / program / library declarations - class and interface type declarations - procedure / function implementations (including qualified TClass.Method n

(path: Path)

def extract_pascal(path: Path) -> dict:
    """Extract units, classes, procedures, uses-imports, and calls from Pascal/Delphi files.

    Produces nodes for:
    - The file itself
    - unit / program / library declarations
    - class and interface type declarations
    - procedure / function implementations (including qualified TClass.Method names)

    Produces edges for:
    - file --contains--> module
    - module --imports--> other file node (via uses clause, resolved to path-based IDs)
    - class --inherits--> base class
    - class/module --contains--> method forward declaration
    - class/module --contains--> procedure/function implementation
    - procedure --calls--> other procedure (within the same file)

    Uses tree-sitter-pascal when available; falls back to a regex-based extractor
    (_extract_pascal_regex) when it isn't installed or fails to parse, so Pascal
    extraction works out of the box without an extra pip install.
    """
    try:
        import tree_sitter_pascal as tspascal
        from tree_sitter import Language, Parser
    except ImportError:
        return _extract_pascal_regex(path)

    try:
        language = Language(tspascal.language())
        parser = Parser(language)
        source = path.read_bytes()
        tree = parser.parse(source)
        root = tree.root_node
    except Exception:
        return _extract_pascal_regex(path)

    stem = _file_stem(path)
    str_path = str(path)
    nodes: list[dict] = []
    edges: list[dict] = []
    seen_ids: set[str] = set()
    proc_bodies: list[tuple[str, Any]] = []

    def _read(node) -> str:  # type: ignore[no-untyped-def]
        return source[node.start_byte:node.end_byte].decode("utf-8", errors="replace")

    def add_node(nid: str, label: str, line: int) -> None:
        if nid not in seen_ids:
            seen_ids.add(nid)
            nodes.append({
                "id": nid, "label": label, "file_type": "code",
                "source_file": str_path, "source_location": f"L{line}",
            })

    def add_edge(
        src: str, tgt: str, relation: str, line: int,
        confidence: str = "EXTRACTED", weight: float = 1.0,
        context: str | None = None,