Skip to main content

py-package-template - Package Template Project Generator

Project description

Python Package Template Project

image Build Status

The py-template-project package allows users to download the contents of this GiHub repository, containing a skeleton Python package project to be used as a template for kick-starting development of any type of Package; destined for upload to PyPI, or just for local install using Pip. The downloaded package includes the following components to aid rapid development without having to spend time cloning existing set-ups from other projects:

  • a minimal file;
  • testing with PyTest;
  • documentation (HTML and PDF) generated using Sphinx with auto-documentation setup;
  • an entry-point that allows the package to execute functions directly from the command line - e.g. to start a server, interact with a user, download a GitHub repository, etc.; and,
  • automated testing and deployment using Travis CI.

A description of how to work with (and modify) each of these components, is provided in more detail in the sections that follow-on below, as well as in the documentation and within the example code bundled with the package.

This is obviously a opinionated view of how a Python package project ought to be structured, based largely on my own experiences and requirements. Where I have needed guidance on this subject, I have leant heavily on the advice given by the Python Packaging Authority (PyPA) and used the excellent Requests and Flask projects as references for 'best practices'.


Install and update using pip:

pip3 install py-template-project

Downloading a Python Package Template Project

To down load the latest version of the Python Package Template project located in this GiHub repository, execute the following command from the command line:

py-package-template install

This will be downloaded to the current directory and will contain the following directory structure:

 |-- docs/
 |-- |-- build_html/
 |-- |-- build_latex/
 |-- |-- source/
 |-- py-pkg/
 |-- |--
 |-- |--
 |-- |--
 |-- |--
 |-- tests/
 |-- |-- test_data/
 |-- |   |-- supply_demand_data.json
 |-- |
 |-- |
 |-- |
 |-- .env
 |-- .gitignore
 |-- Pipfile
 |-- Pipfile.lock

The Python Package Template Project

We now describe the various components of the template project and the workflows associated with it. The template package project contains two modules to get things started:


The module contains sample code for modelling economic supply and demand curves and makes for a useful demonstration of how Python type annotation and interface definition via abstract base classes, can make code easier to read, document and reason about (I am a big fan). The test suite for this module is contained in the tests folder and demonstrates how to get up-and-running with PyTest.

The module is referenced in the file via the entry_points definitions:

    'console_scripts': ['py-package-template=py_pkg.entry_points:main'],

It enables the declared entry point - py_pkg.entry_points.main - to be invoked when py-package-template is called from the command line. This is what enables the template project to be downloaded programmatically (check the code for the full details). This could easily be extended to start a server (e.g. using Flask), or run any other type of script.

Project Dependencies

We use pipenv for managing project dependencies and Python environments (i.e. virtual environments). These dependencies are not to be confused with the package installation dependencies for the package under developement - i.e. those that need to be defined in the install_requires section of All of the direct packages dependencies required to run the project's code (e.g. NumPy for tensors), as well as all the packages used during development (e.g. flake8 for code linting and IPython for interactive console sessions), are described in the Pipfile. Their precise downstream dependencies are crystallised in Pipfile.lock, which is used to guarentee repeatable (i.e. deterministic) builds.

Installing Pipenv

To get started with Pipenv, first of all download it - assuming that there is a 'global' version of Python available on your system and on the PATH, then this can be achieved by running the following command,

pip3 install pipenv

For more information, including advanced configuration options, see the official pipenv documentation.

Installing this Projects' Dependencies

Make sure that you're in the project's root directory (the same one in which Pipfile resides), and then run,

pipenv install --dev

This will install all of the direct project dependencies as well as the development dependencies (the latter a consequence of the --dev flag). To add and remove dependencies as required for your new project, use pipenv install and pipenv uninstall as required, using the --dev flag for development-only dependencies.

Running Python and IPython from the Project's Virtual Environment

In order to open a Python REPL using within an environment that precisely mimics the one the project is being developed with, use Pipenv from the command line as follows,

pipenv run python3

The python3 command could just as well be ipython3.

Automatic Loading of Environment Variables

Pipenv will automatically pick-up and load any environment variables declared in the .env file, located in the package's root directory. For example, adding,


Will enable access to this variable within any Python program, via a call to os.environ['SPARK_HOME']. Note, that if any security credentials are placed here, then this file must be removed from source control - i.e. add .env to the .gitignore file to prevent potential security risks.

Pipenv Shells

Prepending pipenv to every command you want to run within the context of your Pipenv-managed virtual environment, can get (very) tedious. This can be avoided by entering into a Pipenv-managed shell,

pipenv shell

Which is equivalent to 'activating' the virtual environment. Any command will now be executed within the virtual environment. Use exit to leave the shell session.

Running Unit Tests

All test have been written using the PyTest package. Tests are kept in the tests folder and can be run from the command line by - e.g. by invoking,

pipenv run pytest

The test suite is structured as an independent Python package as follows:

 |-- test_data/
 |   |-- supply_demand_data.json

The module is used by PyTest - in this particular instance for loading test data and building objects that will then be used by potentially many other tests. These are referred to as 'fixtures' in PyTest - more details can be found here.

Linting Code

I prefer to use flake8 for style guide enforcement. This can be invoked from the command line by running,

pipenv run flake8 py_pkg

Flake8 could easily be swapped-out for another tool by using Pipenv as described above.

Static Type Checking

We have used the Python type annotation framework, together with the MyPy package, to perform static type checks on the codebase. Analogous to any linter or unit testing framework, MyPy can be run from the command line as follows,

pipenv run python -m mypy py_pkg/*.py

MyPy options for this project can be defined in the mypy.ini file that MyPy will look for by default. For more information on the full set of options, see the mypy documentation.

Examples of type annotation and type checking for library development can be found in the module. This should also be cross-referenced with the improvement to readability (and usability) that this has on package documentation.


The documentation in the docs folder has been built using Sphinx. We have used the default 'quickstart' automatic configuration, which was originally triggered by executing,

pipenv run sphinx-quickstart

The output is based primarily on the Docstrings in the source code, using the autodoc extension within Sphinx (specified during the 'quickstart'). The contents for the entry point into the docs (index.html), is defined in the index.rst file, which itself imports the modules.rst file that lists all of the modules to document. The documentation can be built by running the following command,

pipenv run sphinx-build -b html docs/source docs/build_html

The resulting HTML documentation can be accessed by opening docs/build_html/index.html in a web browser.

My preferred third party theme from Read the Docs has also been used, by installing the sphinx_rtd_theme as a development dependency and modifying docs/source/ as follows:

import sphinx_rtd_theme
html_theme = "sphinx_rtd_theme"
html_theme_path = [sphinx_rtd_theme.get_html_theme_path()]

Creating a PDF Version Using LaTeX

So long as a LaTex distribution is present on your system (e.g. MikTeX for Mac OS X), then it is possible to create a PDF version of the documentation, as well. Start by building the prerequisite LaTex version from the ReStructured Text originals,

pipenv run sphinx-build -b latex docs/source docs/build_latex

Then, navigate to docs/build_latex and run,


Both LaTeX and PDF versions can then be found in docs/build_latex.

Building Deployable Distributions

The recommended (and most pragmatic) way of deploy this package is to build a Python wheel and to then to install it in a fresh virtual environment on the target system. The exact build configuration is determined by the parameters in Note, that this requires that all package dependencies also be specified in the install_requires declaration in, regardless of their entry in Pipfile. For more information on Python packaging refer to the Python Packaging User Guide and the accompanying sample project. To create the Python wheel run,

pipenv run python bdist_wheel

This will create build, py_package_template.egg-info and dist directories - the wheel can be found in the latter. This needs to be copied to the target system (which we are assuming has Python and Pipenv available as a minimum), where it can be installed into a new virtual environment, together with all downstream dependencies, using,

pipenv install path/to/your-package.whl

Automated Testing and Deployment using Travis CI

We have chosen Travis for Continuous Integration (CI) as it integrates very easily with Python and GitHub (where I have granted it access to my public repositories). The configuration details are kept in the .travis.yaml file in the root directory:

ncsudo: required

language: python

  - 3.7-dev

  - pip install pipenv
  - pipenv install --dev

  - pipenv run pytest

  provider: pypi
  user: alexioannides
    secure: my-encrypted-pypi-password
    tags: true
  distributions: bdist_wheel

Briefly, this instructs the Travis build server to:

  1. download, build and install Python 3.7;
  2. install Pipenv
  3. use Pipenv and Pipfile.lock to install all dependencies (dev dependencies are necessary for running PyTest);
  4. run all unit tests using PyTest;
  5. if the tests were run successfully and if we have pushed a new tag (i.e. a release) to the master branch then:
    • build a Python wheel; and,
    • push it to using my PyPI account credentials.

Note that we provide Travis with an encrypted password, that was made using the Travis command line tool (downloaded using HomeBrew on OS X). For more details on this and PyPI deployment more generally see the Travis CI documentation.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for py-package-template, version 0.2.0
Filename, size File type Python version Upload date Hashes
Filename, size py_package_template-0.2.0-py3-none-any.whl (11.2 kB) File type Wheel Python version py3 Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page