Django Shell MCP Server

from __future__ import annotations import pytest import pytest_asyncio from fastmcp import Client from mcp_django.mgmt import core from mcp_django.mgmt.core import CommandErrorResult from mcp_django.server import mcp pytestmark = [pytest.mark.asyncio, pytest.mark.django_db] @pytest_asyncio.fixture(autouse=True) async def initialize_server(): """Initialize the MCP server before tests.""" await mcp.initialize() async def test_management_command_check(): """Test running the 'check' command (safe, read-only).""" async with Client(mcp.server) as client: result = await client.call_tool( "management_execute_command", { "command": "check", }, ) assert result.data is not None assert result.data.status == "success" assert result.data.command == "check" assert result.data.args == [] assert result.data.exception is None # The check command should produce some output to stderr assert ( "System check identified" in result.data.stderr or result.data.stderr == "" ) async def test_management_command_with_args(): """Test running a command with positional arguments.""" async with Client(mcp.server) as client: result = await client.call_tool( "management_execute_command", { "command": "showmigrations", "args": ["tests"], }, ) assert result.data is not None assert result.data.status == "success" assert result.data.command == "showmigrations" assert result.data.args == ["tests"] assert result.data.exception is None async def test_management_command_with_options(): """Test running a command with keyword options.""" async with Client(mcp.server) as client: result = await client.call_tool( "management_execute_command", { "command": "check", "options": {"verbosity": 2}, }, ) assert result.data is not None assert result.data.status == "success" assert result.data.command == "check" assert result.data.exception is None async def test_management_command_with_args_and_options(): """Test running a command with both args and options.""" async with Client(mcp.server) as client: result = await client.call_tool( "management_execute_command", { "command": "showmigrations", "args": ["tests"], "options": {"verbosity": 0}, }, ) assert result.data is not None assert result.data.status == "success" assert result.data.command == "showmigrations" assert result.data.args == ["tests"] assert result.data.exception is None async def test_management_command_invalid_command(): """Test running an invalid/non-existent command.""" async with Client(mcp.server) as client: result = await client.call_tool( "management_execute_command", { "command": "this_command_does_not_exist", }, ) assert result.data is not None assert result.data.status == "error" assert result.data.command == "this_command_does_not_exist" assert result.data.exception is not None assert result.data.exception.type in [ "CommandError", "ManagementUtilityError", ] assert "Unknown command" in result.data.exception.message async def test_management_command_makemigrations_dry_run(): """Test running makemigrations with --dry-run (safe, read-only).""" async with Client(mcp.server) as client: result = await client.call_tool( "management_execute_command", { "command": "makemigrations", "options": {"dry_run": True, "verbosity": 0}, }, ) assert result.data is not None assert result.data.status == "success" assert result.data.command == "makemigrations" assert result.data.args == [] assert result.data.exception is None async def test_management_command_diffsettings(): """Test running diffsettings command (read-only introspection).""" async with Client(mcp.server) as client: result = await client.call_tool( "management_execute_command", { "command": "diffsettings", "options": {"all": True}, }, ) assert result.data is not None assert result.data.status == "success" assert result.data.command == "diffsettings" # Should output settings assert len(result.data.stdout) > 0 async def test_management_command_stdout_capture(): """Test that stdout is properly captured from commands.""" async with Client(mcp.server) as client: result = await client.call_tool( "management_execute_command", { "command": "check", "options": {"verbosity": 2}, }, ) assert result.data is not None assert result.data.status == "success" # With higher verbosity, check should produce output assert isinstance(result.data.stdout, str) assert isinstance(result.data.stderr, str) async def test_management_command_list_in_main_server(): """Test that management_command tool is listed in main server tools.""" async with Client(mcp.server) as client: tools = await client.list_tools() tool_names = [tool.name for tool in tools] assert "management_execute_command" in tool_names # Find the tool and check its metadata mgmt_tool = next( tool for tool in tools if tool.name == "management_execute_command" ) assert mgmt_tool.description is not None assert "management command" in mgmt_tool.description.lower() async def test_list_management_commands(): """Test listing all available management commands.""" async with Client(mcp.server) as client: result = await client.call_tool("management_list_commands", {}) assert result.data is not None assert isinstance(result.data, list) assert len(result.data) > 0 # Check that we have some standard Django commands command_names = [cmd["name"] for cmd in result.data] assert "check" in command_names assert "migrate" in command_names assert "showmigrations" in command_names # Verify structure of command info first_cmd = result.data[0] assert "name" in first_cmd assert "app_name" in first_cmd assert isinstance(first_cmd["name"], str) assert isinstance(first_cmd["app_name"], str) async def test_list_management_commands_includes_custom_commands(): """Test that custom management commands are included in the list.""" async with Client(mcp.server) as client: result = await client.call_tool("management_list_commands", {}) assert result.data is not None command_names = [cmd["name"] for cmd in result.data] # The mcp command from this project should be in the list assert "mcp" in command_names async def test_list_management_commands_sorted(): """Test that management commands are sorted alphabetically.""" async with Client(mcp.server) as client: result = await client.call_tool("management_list_commands", {}) assert result.data is not None command_names = [cmd["name"] for cmd in result.data] # Verify the list is sorted assert command_names == sorted(command_names) async def test_management_command_unexpected_exception(monkeypatch): """Test handling of unexpected exceptions in execute_command tool.""" # Mock the executor to raise an unexpected exception async def mock_execute(*args, **kwargs): raise RuntimeError("Unexpected error in executor") monkeypatch.setattr(core.management_command_executor, "execute", mock_execute) async with Client(mcp.server) as client: # The tool should re-raise unexpected exceptions with pytest.raises(Exception): # Will be wrapped by FastMCP await client.call_tool( "management_execute_command", { "command": "check", }, ) async def test_command_error_result_with_stdout_stderr(): """Test CommandErrorResult __post_init__ with stdout and stderr for coverage.""" # Create a CommandErrorResult with both stdout and stderr # This triggers lines 58 and 60 in mgmt/core.py result = CommandErrorResult( command="test_command", args=("arg1",), options={"opt": "value"}, exception=Exception("Test exception"), stdout="Some stdout output", stderr="Some stderr output", ) # Verify the result was created correctly assert result.command == "test_command" assert result.stdout == "Some stdout output" assert result.stderr == "Some stderr output" assert str(result.exception) == "Test exception"