langchain-mcp-tools 0.0.14__tar.gz → 0.0.16__tar.gz

{langchain_mcp_tools-0.0.14 → langchain_mcp_tools-0.0.16}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: langchain-mcp-tools
-Version: 0.0.14
+Version: 0.0.16
 Summary: Model Context Protocol (MCP) To LangChain Tools Conversion Utility
 Project-URL: Bug Tracker, https://github.com/hideya/langchain-mcp-tools-py/issues
 Project-URL: Source Code, https://github.com/hideya/langchain-mcp-tools-py
@@ -18,8 +18,6 @@ Requires-Dist: mcp>=1.2.0
 Requires-Dist: pyjson5>=1.6.8
 Requires-Dist: pympler>=1.1
 Requires-Dist: python-dotenv>=1.0.1
-Requires-Dist: pytest>=8.3.4
-Requires-Dist: pytest-asyncio>=0.25.2
 Provides-Extra: dev
 Requires-Dist: twine>=6.0.1; extra == "dev"
 
@@ -108,7 +106,7 @@ Currently, only text results of tool calls are supported.
 ## Technical Details
 
 It was very tricky (for me) to get the parallel MCP server initialization
-to work successfully...
+to work, including successful final resource cleanup...
 
 I'm new to Python, so it is very possible that my ignorance is playing
 a big role here...  
@@ -119,7 +117,13 @@ Any comments pointing out something I am missing would be greatly appreciated!
 [(comment here)](https://github.com/hideya/langchain-mcp-tools-ts/issues)
 
 1. Core Challenge:
-   - Async resources management for `stdio_client` and `ClientSession` seems
+
+   A key requirement for parallel initialization is that each server must be
+   initialized in its own dedicated task - there's no way around this as far as
+   I know. However, this poses a challenge when combined with
+   `asynccontextmanager`.
+
+   - Resources management for `stdio_client` and `ClientSession` seems
      to require relying exclusively on `asynccontextmanager` for cleanup,
      with no manual cleanup options
      (based on [the mcp python-sdk impl as of Jan 14, 2025](https://github.com/modelcontextprotocol/python-sdk/tree/99727a9/src/mcp/client))
@@ -131,11 +135,6 @@ Any comments pointing out something I am missing would be greatly appreciated!
 
 2. Solution Strategy:
 
-   A key requirement for parallel initialization is that each server must be
-   initialized in its own dedicated task - there's no way around this as far
-   as I understand. However, this creates a challenge since we also need to
-   maintain long-lived sessions and handle cleanup properly.
-
    The key insight is to keep the initialization tasks alive throughout the
    session lifetime, rather than letting them complete after initialization.
 

{langchain_mcp_tools-0.0.14 → langchain_mcp_tools-0.0.16}/README.md

@@ -83,7 +83,7 @@ Currently, only text results of tool calls are supported.
 ## Technical Details
 
 It was very tricky (for me) to get the parallel MCP server initialization
-to work successfully...
+to work, including successful final resource cleanup...
 
 I'm new to Python, so it is very possible that my ignorance is playing
 a big role here...  
@@ -94,7 +94,13 @@ Any comments pointing out something I am missing would be greatly appreciated!
 [(comment here)](https://github.com/hideya/langchain-mcp-tools-ts/issues)
 
 1. Core Challenge:
-   - Async resources management for `stdio_client` and `ClientSession` seems
+
+   A key requirement for parallel initialization is that each server must be
+   initialized in its own dedicated task - there's no way around this as far as
+   I know. However, this poses a challenge when combined with
+   `asynccontextmanager`.
+
+   - Resources management for `stdio_client` and `ClientSession` seems
      to require relying exclusively on `asynccontextmanager` for cleanup,
      with no manual cleanup options
      (based on [the mcp python-sdk impl as of Jan 14, 2025](https://github.com/modelcontextprotocol/python-sdk/tree/99727a9/src/mcp/client))
@@ -106,11 +112,6 @@ Any comments pointing out something I am missing would be greatly appreciated!
 
 2. Solution Strategy:
 
-   A key requirement for parallel initialization is that each server must be
-   initialized in its own dedicated task - there's no way around this as far
-   as I understand. However, this creates a challenge since we also need to
-   maintain long-lived sessions and handle cleanup properly.
-
    The key insight is to keep the initialization tasks alive throughout the
    session lifetime, rather than letting them complete after initialization.
 

{langchain_mcp_tools-0.0.14 → langchain_mcp_tools-0.0.16}/langchain_mcp_tools/langchain_mcp_tools.py

@@ -37,7 +37,13 @@ require context managers while enabling parallel initialization.
 The key aspects are:
 
 1. Core Challenge:
-   - Async resources management for `stdio_client` and `ClientSession` seems
+
+   A key requirement for parallel initialization is that each server must be
+   initialized in its own dedicated task - there's no way around this as far as
+   I know. However, this poses a challenge when combined with
+   `asynccontextmanager`.
+
+   - Resources management for `stdio_client` and `ClientSession` seems
      to require relying exclusively on `asynccontextmanager` for cleanup,
      with no manual cleanup options
      (based on [the mcp python-sdk impl as of Jan 14, 2025](https://github.com/modelcontextprotocol/python-sdk/tree/99727a9/src/mcp/client))
@@ -48,10 +54,6 @@ The key aspects are:
    - Need to ensure proper cleanup later in the same task that created them
 
 2. Solution Strategy:
-   A key requirement for parallel initialization is that each server must be
-   initialized in its own dedicated task - there's no way around this as far
-   as I understand. However, this creates a challenge since we also need to
-   maintain long-lived sessions and handle cleanup properly.
 
    The key insight is to keep the initialization tasks alive throughout the
    session lifetime, rather than letting them complete after initialization.

{langchain_mcp_tools-0.0.14 → langchain_mcp_tools-0.0.16}/langchain_mcp_tools.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: langchain-mcp-tools
-Version: 0.0.14
+Version: 0.0.16
 Summary: Model Context Protocol (MCP) To LangChain Tools Conversion Utility
 Project-URL: Bug Tracker, https://github.com/hideya/langchain-mcp-tools-py/issues
 Project-URL: Source Code, https://github.com/hideya/langchain-mcp-tools-py
@@ -18,8 +18,6 @@ Requires-Dist: mcp>=1.2.0
 Requires-Dist: pyjson5>=1.6.8
 Requires-Dist: pympler>=1.1
 Requires-Dist: python-dotenv>=1.0.1
-Requires-Dist: pytest>=8.3.4
-Requires-Dist: pytest-asyncio>=0.25.2
 Provides-Extra: dev
 Requires-Dist: twine>=6.0.1; extra == "dev"
 
@@ -108,7 +106,7 @@ Currently, only text results of tool calls are supported.
 ## Technical Details
 
 It was very tricky (for me) to get the parallel MCP server initialization
-to work successfully...
+to work, including successful final resource cleanup...
 
 I'm new to Python, so it is very possible that my ignorance is playing
 a big role here...  
@@ -119,7 +117,13 @@ Any comments pointing out something I am missing would be greatly appreciated!
 [(comment here)](https://github.com/hideya/langchain-mcp-tools-ts/issues)
 
 1. Core Challenge:
-   - Async resources management for `stdio_client` and `ClientSession` seems
+
+   A key requirement for parallel initialization is that each server must be
+   initialized in its own dedicated task - there's no way around this as far as
+   I know. However, this poses a challenge when combined with
+   `asynccontextmanager`.
+
+   - Resources management for `stdio_client` and `ClientSession` seems
      to require relying exclusively on `asynccontextmanager` for cleanup,
      with no manual cleanup options
      (based on [the mcp python-sdk impl as of Jan 14, 2025](https://github.com/modelcontextprotocol/python-sdk/tree/99727a9/src/mcp/client))
@@ -131,11 +135,6 @@ Any comments pointing out something I am missing would be greatly appreciated!
 
 2. Solution Strategy:
 
-   A key requirement for parallel initialization is that each server must be
-   initialized in its own dedicated task - there's no way around this as far
-   as I understand. However, this creates a challenge since we also need to
-   maintain long-lived sessions and handle cleanup properly.
-
    The key insight is to keep the initialization tasks alive throughout the
    session lifetime, rather than letting them complete after initialization.
 

{langchain_mcp_tools-0.0.14 → langchain_mcp_tools-0.0.16}/langchain_mcp_tools.egg-info/SOURCES.txt

@@ -8,5 +8,4 @@ langchain_mcp_tools.egg-info/PKG-INFO
 langchain_mcp_tools.egg-info/SOURCES.txt
 langchain_mcp_tools.egg-info/dependency_links.txt
 langchain_mcp_tools.egg-info/requires.txt
-langchain_mcp_tools.egg-info/top_level.txt
-tests/test_langchain_mcp_tools.py
+langchain_mcp_tools.egg-info/top_level.txt

{langchain_mcp_tools-0.0.14 → langchain_mcp_tools-0.0.16}/langchain_mcp_tools.egg-info/requires.txt

@@ -9,8 +9,6 @@ mcp>=1.2.0
 pyjson5>=1.6.8
 pympler>=1.1
 python-dotenv>=1.0.1
-pytest>=8.3.4
-pytest-asyncio>=0.25.2
 
 [dev]
 twine>=6.0.1

{langchain_mcp_tools-0.0.14 → langchain_mcp_tools-0.0.16}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "langchain-mcp-tools"
-version = "0.0.14"
+version = "0.0.16"
 description = "Model Context Protocol (MCP) To LangChain Tools Conversion Utility"
 readme = "README.md"
 requires-python = ">=3.11"
@@ -16,8 +16,6 @@ dependencies = [
     "pyjson5>=1.6.8",
     "pympler>=1.1",
     "python-dotenv>=1.0.1",
-    "pytest>=8.3.4",
-    "pytest-asyncio>=0.25.2",
 ]
 
 [project.optional-dependencies]

langchain_mcp_tools-0.0.14/tests/test_langchain_mcp_tools.py

@@ -1,163 +0,0 @@
-import pytest
-from unittest.mock import AsyncMock, MagicMock, patch
-from langchain_core.tools import BaseTool
-from langchain_mcp_tools.langchain_mcp_tools import (
-    convert_mcp_to_langchain_tools,
-)
-
-# Fix the asyncio mark warning by installing pytest-asyncio
-pytest_plugins = ('pytest_asyncio',)
-
-
-@pytest.fixture
-def mock_stdio_client():
-    with patch('langchain_mcp_tools.langchain_mcp_tools.stdio_client') as mock:
-        mock.return_value.__aenter__.return_value = (AsyncMock(), AsyncMock())
-        yield mock
-
-
-@pytest.fixture
-def mock_client_session():
-    with patch('langchain_mcp_tools.langchain_mcp_tools.ClientSession') as mock:
-        session = AsyncMock()
-        # Mock the list_tools response
-        session.list_tools.return_value = MagicMock(
-            tools=[
-                MagicMock(
-                    name="tool1",
-                    description="Test tool",
-                    inputSchema={"type": "object", "properties": {}}
-                )
-            ]
-        )
-        mock.return_value.__aenter__.return_value = session
-        yield mock
-
-
-@pytest.mark.asyncio
-async def test_convert_mcp_to_langchain_tools_empty():
-    server_configs = {}
-    tools, cleanup = await convert_mcp_to_langchain_tools(server_configs)
-    assert isinstance(tools, list)
-    assert len(tools) == 0
-    await cleanup()
-
-
-"""
-@pytest.mark.asyncio
-async def test_convert_mcp_to_langchain_tools_invalid_config():
-    server_configs = {"invalid": {"command": "nonexistent"}}
-    with pytest.raises(Exception):
-        await convert_mcp_to_langchain_tools(server_configs)
-"""
-
-
-"""
-@pytest.mark.asyncio
-async def test_convert_single_mcp_success(
-    mock_stdio_client,
-    mock_client_session
-):
-    # Test data
-    server_name = "test_server"
-    server_config = {
-        "command": "test_command",
-        "args": ["--test"],
-        "env": {"TEST_ENV": "value"}
-    }
-    langchain_tools = []
-    ready_event = asyncio.Event()
-    cleanup_event = asyncio.Event()
-
-    # Create task
-    task = asyncio.create_task(
-        convert_single_mcp_to_langchain_tools(
-            server_name,
-            server_config,
-            langchain_tools,
-            ready_event,
-            cleanup_event
-        )
-    )
-
-    # Wait for ready event
-    await asyncio.wait_for(ready_event.wait(), timeout=1.0)
-
-    # Verify tools were created
-    assert len(langchain_tools) == 1
-    assert isinstance(langchain_tools[0], BaseTool)
-    assert langchain_tools[0].name == "tool1"
-
-    # Trigger cleanup
-    cleanup_event.set()
-    await task
-"""
-
-
-@pytest.mark.asyncio
-async def test_convert_mcp_to_langchain_tools_multiple_servers(
-    mock_stdio_client,
-    mock_client_session
-):
-    server_configs = {
-        "server1": {"command": "cmd1", "args": []},
-        "server2": {"command": "cmd2", "args": []}
-    }
-
-    tools, cleanup = await convert_mcp_to_langchain_tools(server_configs)
-
-    # Verify correct number of tools created
-    assert len(tools) == 2  # One tool per server
-    assert all(isinstance(tool, BaseTool) for tool in tools)
-
-    # Test cleanup
-    await cleanup()
-
-
-"""
-@pytest.mark.asyncio
-async def test_tool_execution(mock_stdio_client, mock_client_session):
-    server_configs = {
-        "test_server": {"command": "test", "args": []}
-    }
-
-    # Mock the tool execution response
-    session = mock_client_session.return_value.__aenter__.return_value
-    session.call_tool.return_value = MagicMock(
-        isError=False,
-        content={"result": "success"}
-    )
-
-    tools, cleanup = await convert_mcp_to_langchain_tools(server_configs)
-
-    # Test tool execution
-    result = await tools[0]._arun(test_param="value")
-    assert result == {"result": "success"}
-
-    # Verify tool was called with correct parameters
-    session.call_tool.assert_called_once_with("tool1", {"test_param": "value"})
-
-    await cleanup()
-"""
-
-
-@pytest.mark.asyncio
-async def test_tool_execution_error(mock_stdio_client, mock_client_session):
-    server_configs = {
-        "test_server": {"command": "test", "args": []}
-    }
-
-    # Mock error response
-    session = mock_client_session.return_value.__aenter__.return_value
-    session.call_tool.return_value = MagicMock(
-        isError=True,
-        content="Error message"
-    )
-
-    tools, cleanup = await convert_mcp_to_langchain_tools(server_configs)
-
-    # Test tool execution error
-    with pytest.raises(Exception):
-        await tools[0]._arun(test_param="value")
-
-    await cleanup()