This directory contains the integration tests. These are written in Python because the NUnit runners don't seem to consistently log both external process output and Console.WriteLine messages, making debugging a lot harder.
Please note that these integration tests are not meant to show no errors/warnings at all. Because of the differences between C# and Python, it is sometimes necessary to perform conversions which are invalid according to common PEPs, as long as editors can still read the stubs and provide autocomplete and the like properly.
Before running integration_tests.py, make sure pandas and matplotlib are installed in your Python environment and the following commands are available on your PATH:
git, to clone/pull the latest version of Leandotnet, to run the generatorpyright, to check the generated stubs using the Pyright type checker used in Pylance (npm i -g pyrightoryarn global add pyright)
During development it is also useful to check the generated stubs in PyCharm and in VS Code with the Pylance extension. Good files to open and see if the stubs provide correct information can be found in QuantConnect/Lean/Algorithm.Python.