Python development environment with Windows + Python + PipEnv + Visual Studio Code

I wrote the following article a long time ago, but it seems that the situation has changed with the update, so Update the content and repost.

Python development environment with Windows + Python 3.6 + PipEnv + Visual Studio Code

■ Environment

• Windows 10
• Python version 3.8.2
• PipEnv version 2018.11.26

■ Python installation

Download from Python official website. Here, python-3.8.2-amd64.exe is used.

Run the installer to install Python. Please note the following.

• Install pip (-> check "pip" on page 2)
• Add Python to environment variables (-> Check "Add Python to environment variables" on page 3)

■ Installing PipEnv

Start a command prompt and do the following:

Install PipEnv

pip install pipenv

● Add environment variables

By default, PipEnv stores installed packages under your home. By adding the following environment variable, it will be stored in the project directory.

• PIPENV_VENV_IN_PROJECT : true

■ VS Code initial settings

● Extensions

Add the following extensions.

• Python（ms-python.python）

● User settings.json settings

The activate command may not work well with PowerShell (?) It seems that it does not work well again with the workspace setting (?), So Add the following in your custom settings.json to make the command prompt the default.

[User home]\AppData\Roaming\Code\User\settings.json

{
    "terminal.integrated.shell.windows": "C:/WINDOWS/System32/cmd.exe",
}

■ From VS Code project creation to execution

● Create project folder & create PipEnv environment

Cut a directory of your choice and open it with VS Code. Here, it is "D: \ Temp \ python-test".

Open a terminal in View-> Integrated Terminal and run the PipEnv command. Change the arguments according to the version of Python you are using.

Create PipEnv environment

pipenv --python 3.8

When executed, the following will be created directly under the project.

• .venv
• PipFile

● Workspace settings.json settings

Set the path for the virtual environment, formatter, and Lint.

json:[Workspace folder]\.vscode\settings.json

{
    //Activate the Python environment in the terminal when the extension loads.
    "python.terminal.activateEnvInCurrentTerminal": true,
    //Virtual environment path. Specify the created virtual environment.
    "python.venvPath": "{$workspaceFolder}/.venv",
    "python.autoComplete.extraPaths": [
        "{$workspaceFolder}/.venv/Lib/site-packages",
    ],
    //Formatter settings. Specify autopep8.
    "python.formatting.provider": "autopep8",
    "python.jediEnabled": false,
    //Lint settings. Enable flake8 and mypy.
    "python.linting.flake8Enabled": true,
    "python.linting.mypyEnabled": true,
    "python.linting.pylintEnabled": false,
    //The following is your choice.
    "editor.formatOnSave": true,
    "python.autoComplete.addBrackets": true,
}

● Package installation

The package is installed in the virtual environment created by using the pipenv command on the terminal. Install the following packages for the code format Lint. Add the --dev option for use only in the development environment.

• autopep8
• flake8
• mypy

Installation of autopep8, flake8, mypy

pipenv install autopep8 flake8 mypy --dev

Here, in addition to the above package, the following is added for the sample source.

• numpy
• matplotlib

Installation of numpy and matplotlib

pipenv install numpy matplotlib

● Execution

Create a Python source file as you like.

test.py

import numpy as np
import matplotlib.pyplot as plt

x = np.linspace(-np.pi, np.pi)
y = np.sin(x)
plt.plot(x, y)
plt.show()

If the settings are correct, document viewing, code completion, formatter, and Lint should work.

If you want to run it, open the Run panel and Select "Create a launch.json file"-> "Python File".

.Vscode / launch.json will be created without permission, and the execution configuration for Python will be set.

After that, open the source to be executed and press the F5 key to execute it.

If you want to specify the execution argument, add the setting of "" args "" to .vscode / launch.json.

        {
            "name": "Python: Current File",
            
            "args": [
                "hoge",
                "fuga",
                "piyo"
            ]
        },

■ .gitignore

/.venv
/.mypy_cache
/Pipfile.lock
/tmp
/**/__pycache__