2. Dependencies

2. Dependency Management

Explicitly declare and isolate dependencies

Pip & Virtualenv

Python comes pre-installed with PIP (Python Package Installer) for declaration and Virtualenv for Isolation. PIP allows installation of packages from Python Package Index (PyPI)

Create a virtual environment to manage dependencies

python -m venv env

Activate the Virtual Environmnet

on Windows: env\Scripts\activate
On macOS/Linux:source env/bin/activate

Install Dependencies

by listing the pacakges: pip install numpy scipy pandas scikit-learn matplotlib seaborn
by using requirement file: pip install -r requirements.txt
from local git repo: pip install git+file:///path/to/your/package#egg=package-name
from remote git repo optionally specifying branch: pip install git+ssh://[email protected]:Username/Project.git@master

To work with specific python version you'd have to call pip from the specific python version installed to the system eg:

python3.12 -m pip install numpy

Anaconda/Miniconda

Many developer also prefer to use Anaconda or minimal version miniconda for package and environment management. Ananconda distribution comes with many preinstalled package and many others can be installed from Anaconda Please follow and install the instructions mentioned in their site. If you installed everything correctly you'll be able to start conda environment by typing conda init in the terminal.

Creating a New Environment, with a specific python version

conda create -n myenv python=3.8

Activate your environment with: conda activate myenv

Deactivate your environment with: conda deactivate

Install pacakges into environment, you can also specify channels or install a local file

conda install numpy pandas matplotlib
conda install -c conda-forge scikit-learn
conda install /path/to/package-file.tar.bz2

Managing multiple environments

conda env list
conda env remove -n myenv
conda update conda

Export your environment to YAML file for sharing or create from an existing YAML file

conda env export > environment.yml
conda env create -f environment.yml

Poetry

Poetry is a mordern packaging and dependency management tool that aims to simplify the process of managing project dependencies and publishing packages to PyPI. It uses a single file, pyproject.toml, to manage both project metadata and dependencies, making it easier to maintain and understand the project structure.

Installation: pip install poetry --user --upgrade --pre

Creating a new Project

poetry new poetry-demo-new
poetry new --src poetry-demo-new-src

Initializing Existing project

cd poetry-demo-init
poetry init

Poetry either uses your configured virtualenvs or creates its own to always be isolated from your system.

cd poetry-demo-transition
poetry init
poetry add `cat requirements.txt`
poetry add --dev `cat dev_requirements.txt`
poetry export -f requirements.txt --output requirements.txt
poetry show --tree

Run Commands within poetry virtual environment

poetry run pytest

Ability to work within an activated virtual env

poetry shell

Environment Commands

poetry env info
poetry env list
poetry env use 3.8
poetry env remove 3.7

You need to have specific version of python installed in your system in order for poetry to use it

Other useful commands

poetry search dynaconf
poetry add dynaconf
poetry add coverage --dev
poetry update
poetry update pytest
poetry remove dynaconf
poetry remove coverage --dev
poetry install
poetry install --no-dev
poetry export --without-hashes -o requirements.txt
poetry export --dev --without-hashes -o dev_requirements.txt

Easily build and package your projects with a single command. Supports source distribution and wheels.

poetry build

Make your work known by publishing it to PyPI. You can also publish on private repositories.

poetry publish