chore: Add additional test cases to cover tool invocation and better docstring validation.

Author: anubhav756

packages/toolbox-core/tests/test_tools.py

@@ -13,7 +13,115 @@
 # limitations under the License.
 
 
-from toolbox_core.tool import create_docstring
+from typing import AsyncGenerator
+
+import pytest
+import pytest_asyncio
+from aiohttp import ClientSession
+from aioresponses import aioresponses
+from pydantic import ValidationError
+
+from toolbox_core.protocol import ParameterSchema
+from toolbox_core.tool import ToolboxTool, create_docstring
+
+TEST_BASE_URL = "http://toolbox.example.com"
+TEST_TOOL_NAME = "sample_tool"
+
+
+@pytest.fixture
+def sample_tool_params() -> list[ParameterSchema]:
+    """Parameters for the sample tool."""
+    return [
+        ParameterSchema(
+            name="message", type="string", description="A message to process"
+        ),
+        ParameterSchema(name="count", type="integer", description="A number"),
+    ]
+
+
+@pytest.fixture
+def sample_tool_description() -> str:
+    """Description for the sample tool."""
+    return "A sample tool that processes a message and a count."
+
+
+@pytest_asyncio.fixture
+async def http_session() -> AsyncGenerator[ClientSession, None]:
+    """Provides an aiohttp ClientSession that is closed after the test."""
+    async with ClientSession() as session:
+        yield session
+
+
+def test_create_docstring_one_param_real_schema():
+    """
+    Tests create_docstring with one real ParameterSchema instance.
+    """
+    description = "This tool does one thing."
+    params = [
+        ParameterSchema(
+            name="input_file", type="string", description="Path to the input file."
+        )
+    ]
+
+    result_docstring = create_docstring(description, params)
+
+    expected_docstring = (
+        "This tool does one thing.\n\n"
+        "Args:\n"
+        "    input_file (str): Path to the input file."
+    )
+
+    assert result_docstring == expected_docstring
+
+
+def test_create_docstring_multiple_params_real_schema():
+    """
+    Tests create_docstring with multiple real ParameterSchema instances.
+    """
+    description = "This tool does multiple things."
+    params = [
+        ParameterSchema(name="query", type="string", description="The search query."),
+        ParameterSchema(
+            name="max_results", type="integer", description="Maximum results to return."
+        ),
+        ParameterSchema(
+            name="verbose", type="boolean", description="Enable verbose output."
+        ),
+    ]
+
+    result_docstring = create_docstring(description, params)
+
+    expected_docstring = (
+        "This tool does multiple things.\n\n"
+        "Args:\n"
+        "    query (str): The search query.\n"
+        "    max_results (int): Maximum results to return.\n"
+        "    verbose (bool): Enable verbose output."
+    )
+
+    assert result_docstring == expected_docstring
+
+
+def test_create_docstring_no_description_real_schema():
+    """
+    Tests create_docstring with empty description and one real ParameterSchema.
+    """
+    description = ""
+    params = [
+        ParameterSchema(
+            name="config_id", type="string", description="The ID of the configuration."
+        )
+    ]
+
+    result_docstring = create_docstring(description, params)
+
+    expected_docstring = (
+        "\n\nArgs:\n" "    config_id (str): The ID of the configuration."
+    )
+
+    assert result_docstring == expected_docstring
+    assert result_docstring.startswith("\n\nArgs:")
+    assert "config_id (str): The ID of the configuration." in result_docstring
 
 
 def test_create_docstring_no_params():
@@ -27,3 +135,91 @@ def test_create_docstring_no_params():
 
     assert result_docstring == description
     assert "\n\nArgs:" not in result_docstring
+
+
+@pytest.mark.asyncio
+async def test_tool_creation_callable_and_run(
+    http_session: ClientSession,
+    sample_tool_params: list[ParameterSchema],
+    sample_tool_description: str,
+):
+    """
+    Tests creating a ToolboxTool, checks callability, and simulates a run.
+    """
+    tool_name = TEST_TOOL_NAME
+    base_url = TEST_BASE_URL
+    invoke_url = f"{base_url}/api/tool/{tool_name}/invoke"
+
+    input_args = {"message": "hello world", "count": 5}
+    expected_payload = input_args.copy()
+    mock_server_response_body = {"result": "Processed: hello world (5 times)"}
+    expected_tool_result = mock_server_response_body["result"]
+
+    with aioresponses() as m:
+        m.post(invoke_url, status=200, payload=mock_server_response_body)
+
+        tool_instance = ToolboxTool(
+            session=http_session,
+            base_url=base_url,
+            name=tool_name,
+            description=sample_tool_description,
+            params=sample_tool_params,
+            required_authn_params={},
+            auth_service_token_getters={},
+            bound_params={},
+        )
+
+        assert callable(tool_instance), "ToolboxTool instance should be callable"
+
+        assert "message" in tool_instance.__signature__.parameters
+        assert "count" in tool_instance.__signature__.parameters
+        assert tool_instance.__signature__.parameters["message"].annotation == str
+        assert tool_instance.__signature__.parameters["count"].annotation == int
+
+        actual_result = await tool_instance("hello world", 5)
+
+        assert actual_result == expected_tool_result
+
+        m.assert_called_once_with(
+            invoke_url, method="POST", json=expected_payload, headers={}
+        )
+
+
+@pytest.mark.asyncio
+async def test_tool_run_with_pydantic_validation_error(
+    http_session: ClientSession,
+    sample_tool_params: list[ParameterSchema],
+    sample_tool_description: str,
+):
+    """
+    Tests that calling the tool with incorrect argument types raises an error
+    due to Pydantic validation *before* making an HTTP request.
+    """
+    tool_name = TEST_TOOL_NAME
+    base_url = TEST_BASE_URL
+    invoke_url = f"{base_url}/api/tool/{tool_name}/invoke"
+
+    with aioresponses() as m:
+        m.post(invoke_url, status=200, payload={"result": "Should not be called"})
+
+        tool_instance = ToolboxTool(
+            session=http_session,
+            base_url=base_url,
+            name=tool_name,
+            description=sample_tool_description,
+            params=sample_tool_params,
+            required_authn_params={},
+            auth_service_token_getters={},
+            bound_params={},
+        )
+
+        assert callable(tool_instance)
+
+        with pytest.raises(ValidationError) as exc_info:
+            await tool_instance(message="hello", count="not-a-number")
+
+        assert (
+            "1 validation error for sample_tool\ncount\n  Input should be a valid integer, unable to parse string as an integer [type=int_parsing, input_value='not-a-number', input_type=str]\n    For further information visit https://errors.pydantic.dev/2.11/v/int_parsing"
+            in str(exc_info.value)
+        )
+        m.assert_not_called()