---

Cookiecutter PyGitRepo

cookiecutter-pygitrepo is a cookiecutter template for Python / pip installable / AWS Project.

It is to replace my other project template library pygitrepo.

Features

Create and Remove virtualenv:

# print useful information of your project
make info

# create virtualenv
make up

# remove virtualenv
make remove

Install your library and Run Test:

# run test with pytest
make test

# run code coverage test
make cov

# run test with multiple python version
make tox

Automatical Integration with Travis-CI and Codecov. Configuration free.

Auto format your source code and test code style into pep8:

make reformat

Build Document with auto-generated API reference:

# build sphinx html document site
make build-doc

# open your doc in browser
make view-doc

# deploy your doc website to s3
make deploy-doc

Want to publish to Python Package Index ?, Let's do:

make publish

Run jupyter nootbook:

make notebook

You can easily choose to generate the following tools for your project.

Python Library Project

DevOps Project (Optional, Required if a AWS Project)

It provides a Multi Environment config management tools out-of-the-box. It allows you to:

1. define complex derivable config value logic.
2. switch between configs between multiple environments like dev | test | prod.
3. read configs or sensitive data from any where securely. Like json or xml file from your local computer, from remote server, AWS Parameter store, AWS Secret Manager, database. Since it's python, you can do anything.
4. inject configs value to any tools or systems. Like your web application, Terraform, Ansible, VM, Docker, anything.

Magic happens in ./config directory.

AWS Project (Optional)

AWS Lambda Project

It provides bin tools allows you to easily to:

• package your Lambda Source Code
• build and upload AWS Layer Code

AWS CloudFormation Project

It provides frameworks allows you to dynamically construct CloudFormation template and easily deploy it to multi-environment.

AWS Lambda Relative Features

It is a AWS Lambda function project? Maybe you want to build a AWS Lambda Runtime compatible (Linux) deployment package (require docker):

make lbd-build-everything

You can simulate lambda invoke locally with custom event data like this:

bash ./bin/lbd/invoke my_func ./lbd-test-event.json

layers can minimize your build / deploy process to fewer second. You can build the dependencies once, and deploy it everywhere. Just do:

make lbd-build-layer
make lbd-upload-everything
make lbd-deploy-layer

Continues deployment is easy, just do this when it passed the Continues Integration. It use severless framework to manage aws infrastructure and configurations:

make lbd-deploy-all-func

Usage: Generate Project Skeleton

cd <to-any-dir-for-example-your-github-dir>
git clone https://github.com/MacHu-GWU/cookiecutter-pygitrepo.git
python ./cookiecutter-pygitrepo/run_cookiecutter_pygitrepo.py

Note:

since it use extra_context to inject runtime data such as today's date to the main context, this won't work properly:

cookiecutter https://github.com/MacHu-GWU/cookiecutter-pygitrepo.git

Description

notation:

• <tmp>: indicate that this directory / file are temporarily generated, and should not be included in the repo.

<repo_name>
|-- .github
|-- .circleci
|-- bin
    |-- py                  # python environment bin tools and shell scripts
    |-- lbd                 # aws lambda bin tools and shell scripts
    |-- source              # common shell scripts functions
    |-- settings.sh
|-- build                   # <tmp>
    |-- lambda              # <tmp> for aws lambda function deployment
        |-- source.zip      # <tmp> source code only
        |-- deploy-pkg.zip  # <tmp> source code + dependencies
        |-- layer.zip       # <tmp> dependencies layer
        |-- site-packages   # <tmp> local lambda runtime sandbox, will be used in container
|-- docs
    |-- build               # <tmp>
        |-- html            # <tmp>
    |-- source
        |-- _static
            |-- css
            |-- js
            |-- .custom-style.rst
            |-- xxx-favicon.ico
            |-- xxx-logo.png
        |-- conf.py
        |-- index.rst
        |-- release-history.rst
    |-- make.bat
    |-- Makefile
|-- <package_name>
    |-- docs
        |-- __init__.py
    |-- handlers
        |-- __init__.py
        |-- lambda_function1.py
        |-- lambda_function2.py
        |-- ...
    |-- __init__.py
    |-- cli.py
    |-- _version.py
|-- tests
    |-- all.py              # invoke all test from python script
    |-- test_xxx.py
|-- .coveragerc
|-- codecov.yml
|-- .gitattributes
|-- .gitignore
|-- .travis.yml
|-- AUTHORS.rst
|-- fix_code_style.py
|-- LICENSE.txt
|-- Makefile                # quick access
|-- MANIFEST.in
|-- README.rst
|-- readthedocs.yml
|-- release-history.rst
|-- requirements.txt
|-- requirements-dev.txt
|-- requirements-doc.txt
|-- requirements-test.txt
|-- serverless.yml          # aws lambda configuration
|-- setup.py
|-- tox.ini

Dev Guide

To test your change to {{ cookiecutter.repo_name }}

1. Run dev.py, generate learn_awslambda-project repo.
2. Test with make xxx command.

Release Strategy

A new version will be released to a release branch every 3 - 6 months.

The master branch represent the latest development version.

All release branches are production ready.

To use specific version, just clone the specific release branch and run python ./init_repo.py