Python project and environment setup

Python projects use packages to extend the base language set. Both virtualenv and pipenv provide a way to manage the packages in a project

Background
Pre-requisites
- PIPENV setup
- VIRTUALENV setup
Project Setup (Linux)

Background

virtualenv creates a sandbox inside a project folder which contains the package cache
- Packages are securely encapsulated inside the project folder
- Multiple copies of a project therefore can exist side-by-side in separate folders, where packages can be independently verified
pipenv stores a cache of packages which a project can use, using virtualenv under the covers to manage project package dependencies
- Using pipenv locally ensures that only the current user can access the packages (Recommended)
- Using pipenv globally provides system-wide access to the libraries
- Creates a Pipfile which contains a list of the Python project requirements
- Creates a Pipfile.lock file which specifies the versions of each package that should be used

Pre-requisites

Install Python 3 globally sudo apt install python3
Install pip3 globally sudo apt install pip3

PIPENV setup

Install pipenv locally pip install --user pipenv

As the user bin path to python is not accessible globally, and not defined by the installation, add the following into the ~/.profile file to declare the path to python

 # add Python pip and pipenv paths
 PYTHON_BIN_PATH="$(python3 -m site --user-base)/bin"
 if [ -d "$PYTHON_BIN_PATH" ] ; then
   PATH="$PATH:$PYTHON_BIN_PATH"
 fi

VIRTUALENV setup

Install virtualenv globally sudo pip install virtualenv, which allows other host users to use virtualenv also

Project Setup (Linux)

Create the project directory

Create the project directory mkdir <project_dir>
Change to the project directory cd <project_dir>

Create GIT repository

Create a local git repository git init
Create a git ignore file touch .gitignore

Managing Project Packages via virtualenv

These operations are on the project, and the command should only be run from within the project root folder.

Operation	Command
Get version	`virtualenv --version`
Create virtual environment venv (or any other name)	`virtualenv venv`
Activating a virtual environment, (use when not activated already)	`source venv/bin/activate`
Leave the virtual environment	`deactivate`
Delete a virtual environment	`rm -rf venv`

Managing Project Packages via pipenv

These operations are on the project, and may modify the cache. Only run these commands from within the project root folder.

Operation	Command
Create a pipenv enabled environment	`pipenv install`
Open a shell for the environment	`pipenv shell`
Exit the shell	`exit`
Add a package	`pipenv install <package_name>`
Remove a package	`pipenv uninstall <package_name>`
Update a package	`pipenv update <package_name>`
Run a project outside the environment shell	`pipenv run python3 <python_file.py>`
Run a project inside the environment shell	`python3 <python_file.py>`
Clean the package cache, removing packages found which are not in the Pipfile.lock	`pipenv clean`

Environment consistency with virtualenv or pipenv

Operation	Command
Freeze packages creates a file which lists the packages inside a local virtual environment	`pip freeze > requirements.txt`
Install requirements from a freeze list file	`pip install -r requirements.txt`

Project Hierarchy

If your project contains subfolders which contain python scripts which cross reference other folders, accessing those folders requires a root python folder.

In the root folder create an empty init file touch __init__.py