drkostas / starter

A starter template for Python packages

Starter

Table of Contents

• About
• Getting Started
  • Prerequisites
• Installing
• Todo
• Built With
• License
• Acknowledgments

About

A cookiecutter template for Python packages.

Getting Started

These instructions will help you generate a new python project/package from the cookiecutter template.

There are 2 ways to use this template:

1. Just open a command line and run:

cookiecutter https://github.com/drkostas/starter.git

You have to have cookiecutter installed first of course. If you don't, run: pip install cookiecutter~=1.7.2 first

2. Clone this repository and follow the next steps

Prerequisites

You need to have a machine with Python > 3.6 and any Bash based shell (e.g. zsh) installed.

$ python3.8 -V
Python 3.8.5

$ echo $SHELL
/usr/bin/zsh

Installing

All the installation steps are being handled by the Makefile.

Just execute the following command and fill in the project variables that will be requested:

$ make install

To delete the cookiecutter directory run:

$ make clean

TODO

Read the TODO to see the current task list.

Built With

• Cookie - A command-line utility that creates projects from cookiecutters (project templates)

License

This project is licensed under the Apache License - see the LICENSE file for details.

Acknowledgments

• Thanks to PurpleBooth for the README template

starter's Issues

Create setup guide with things to do when loading this template

• Create script that takes a name argument and updates the names automatically (Cookiecutter?)
• Setup guide
• Abstractize the repository as much as possible

Add enhancements to the FancyLog package

• Option to log new lines
• Option to log in bold, italic, and strikethrough
• Option to override the customization of a logger for only one print

Personalize this project

• Merge .gitignore files
• Add argparse instead of argh
• Add your Configuration class for yml support
• Add sample unit tests using the unittest library
• Merge the 2 setup.py files
• Add your Makefile
• Add .circleci file
• Modify the cli
• Add timeit
• Add your colorized log class
• Create Example class containing AbstractClass
• Create a Readme file
• Recalibrate all of the above appropriately

Improve abstraction and fine-tune readme, makefile

• Change Makefile to echo the name of env after install
• Put quote in the env file values. Do the same for the packages
• Change the Makefile based on the addition you made on other projects
• Make top-level inits more abstract (for cookiecutter)
• Configurable python_requires and classifiers in setup.py
• Change the readme and makefile to use env=venv or conda
• Change the acknowledgments

Make timeit a separate package

• Make timeit a separate module to be imported by others, because if you import them
  from main, and the main imports them, you get a circular import error
• Convert timeit wrapper to dual decorator and contextmanager

Upload packages to pypi and import them

Create `profileit` decorator/contextmanager

Add db, cloudstore packages from the old template project

Localize imports and improve timeit, logger

Modify timeit to skip more parts of the code when skip=True