chunk_python_code.py

import ast
from langchain.schema import Document 

import ast
from langchain.schema import Document 

def chunk_python_code_with_metadata(source_code, reference):
    """
    Entry point method to process the Python file.
    It invokes the iterate_ast function.
    """
    documents = []
    print(f"Processing file: {reference}")

    iterate_ast(source_code, documents, reference)

    # Determine usage based on the reference path
    if reference.startswith("kadi_apy/lib/"):
        usage = "library"
    elif reference.startswith("kadi_apy/cli/"):
        usage = "cli_library"
    elif reference.startswith("doc/"):
        usage = "doc"
    else:
        usage = "undefined"

    # Add metadata for usage to all documents
    for doc in documents:
        doc.metadata["reference"] = reference
        doc.metadata["usage"] = usage  # Add the determined usage metadata
        print(doc)
    return documents


def iterate_ast(source_code, documents, reference):
    """
    Parses the AST of the given Python file and delegates
    handling to specific methods based on node types.
    """
    # Parse the source code into an abstract syntax tree (AST)
    tree = ast.parse(source_code, filename=reference)

    # Gather all top-level imports for later use
    imports_dict = extract_imports(tree)

    first_level_nodes = list(ast.iter_child_nodes(tree))

    # Check if there are no first-level nodes
    if not first_level_nodes:
        handle_no_first_level_node_found(documents, source_code, imports_dict, reference)
        return

    all_imports = all(isinstance(node, (ast.Import, ast.ImportFrom)) for node in first_level_nodes)
    if all_imports:
        handle_first_level_imports_only(documents, source_code, imports_dict, reference)
    
    # Iterate over first-level nodes
    for first_level_node in ast.iter_child_nodes(tree):
        if isinstance(first_level_node, ast.ClassDef):
            handle_first_level_class(first_level_node, documents, source_code, imports_dict)
        elif isinstance(first_level_node, ast.FunctionDef):
            handle_first_level_func(first_level_node, documents, source_code, imports_dict)
        elif isinstance(first_level_node, ast.Assign):
            handle_first_level_assign(first_level_node, documents, source_code, imports_dict)


def handle_first_level_imports_only(documents, source_code, imports_dict, reference):
    """
    Handles cases where the first-level nodes are only imports.
    """
    if reference.endswith("__init__.py"):
        type = "__init__-file"
    else:
        type = "undefined"

    # Create metadata with "none" if imports are empty
    metadata = {"type": type, "imports": "none" if not imports_dict else imports_dict}

    # Create and store a Document with the full source code
    doc = Document(
        page_content=source_code,
        metadata=metadata
    )
    documents.append(doc)


def extract_imports(tree):
    """
    Extracts all import statements from the AST tree and organizes them
    into a dictionary keyed by their fully qualified names for later analysis.
    """
    imports_dict = {}
    for node in ast.walk(tree):
        if isinstance(node, ast.Import):
            for alias in node.names:
                imports_dict[alias.name] = alias.name
        elif isinstance(node, ast.ImportFrom):
            module = node.module if node.module else ""
            for alias in node.names:
                full_name = f"{module}.{alias.name}" if module else alias.name
                imports_dict[alias.name] = full_name
    return imports_dict


def analyze_imports(node, imports_dict):
    """
    Analyzes the given node's body and signature to find relevant imports.
    """
    relevant_imports = set()
    for sub_node in ast.walk(node):
        if isinstance(sub_node, ast.Name) and sub_node.id in imports_dict:
            relevant_imports.add(imports_dict[sub_node.id])
    return list(relevant_imports)


def handle_no_first_level_node_found(documents, source_code, imports_dict, reference):
    """
    Handles cases where no top-level nodes are found in the AST.
    """
    if reference.endswith("__init__.py"):
        type = "__init__-file"
    else:
        type = "undefined"

    # Create metadata with "none" if imports are empty
    metadata = {"type": type, "imports": "none" if not imports_dict else imports_dict}

    # Create and store a Document with the full source code
    doc = Document(
        page_content=source_code,
        metadata=metadata
    )
    documents.append(doc)


def handle_first_level_assign(assign_node, documents, source_code, imports_dict):
    """
    Handles assignment statements at the first level of the AST.
    """
    assign_start_line = assign_node.lineno
    assign_end_line = assign_node.end_lineno
    assign_source = '\n'.join(source_code.splitlines()[assign_start_line-1:assign_end_line])

    # Extract relevant imports
    assign_imports = analyze_imports(assign_node, imports_dict)

    # Create metadata with "none" if imports are empty
    metadata = {"type": "Assign", "imports": "none" if not assign_imports else assign_imports}

    # Create and store Document for the assignment
    doc = Document(
        page_content=assign_source,
        metadata=metadata
    )
    documents.append(doc)


def handle_first_level_class(class_node, documents, source_code, imports_dict):
    """
    Handles classes at the first level of the AST.
    """
    class_start_line = class_node.lineno
    class_body_lines = [child.lineno for child in class_node.body if isinstance(child, ast.FunctionDef)]
    class_end_line = min(class_body_lines, default=class_node.end_lineno) - 1
    class_source = '\n'.join(source_code.splitlines()[class_start_line-1:class_end_line])

    # Extract relevant imports
    class_imports = analyze_imports(class_node, imports_dict)

    # Create metadata with "none" if imports are empty
    metadata = {
        "type": "class",
        "class": class_node.name,
        "visibility": "public",
        "imports": "none" if not class_imports else class_imports
    }

    # Create and store Document for the class
    doc = Document(
        page_content=class_source,
        metadata=metadata
    )
    documents.append(doc)

    # Handle methods within the class
    for second_level_node in ast.iter_child_nodes(class_node):
        if isinstance(second_level_node, ast.FunctionDef):
            method_start_line = (
                second_level_node.decorator_list[0].lineno
                if second_level_node.decorator_list else second_level_node.lineno
            )
            method_end_line = second_level_node.end_lineno
            method_source = '\n'.join(source_code.splitlines()[method_start_line-1:method_end_line])

            visibility = "internal" if second_level_node.name.startswith("_") else "public"
            method_imports = analyze_imports(second_level_node, imports_dict)

            doc = Document(
                page_content=method_source,
                metadata={
                    "type": "method",
                    "method": second_level_node.name,
                    "visibility": visibility,
                    "imports": "none" if not method_imports else method_imports,
                    "class": class_node.name
                }
            )
            documents.append(doc)


def handle_first_level_func(function_node, documents, source_code, imports_dict):
    """
    Handles functions at the first level of the AST.
    """
    function_start_line = (
        function_node.decorator_list[0].lineno
        if function_node.decorator_list else function_node.lineno
    )
    function_end_line = function_node.end_lineno
    function_source = '\n'.join(source_code.splitlines()[function_start_line-1:function_end_line])

    # Extract relevant imports
    function_imports = analyze_imports(function_node, imports_dict)

    visibility = "internal" if function_node.name.startswith("_") else "public"

    is_command = any(
        decorator.id == "apy_command"
        for decorator in function_node.decorator_list
        if hasattr(decorator, "id")
    )

    metadata = {
        "type": "command" if is_command else "function",
        "visibility": visibility,
        "imports": "none" if not function_imports else function_imports
    }
    if is_command:
        metadata["command"] = function_node.name
    else:
        metadata["method"] = function_node.name

    doc = Document(
        page_content=function_source,
        metadata=metadata
    )
    documents.append(doc)