PyQuick

This is a python app/lib generator that creates a 'empty' python app/lib. And the app/lib 100% uses docker as base for development, test and build. Of course, this project itself 100% on docker :D

Demo

Generate A New Empty Python App

Why PyQuick Created?

Python's Pain Points

• Python is interpreted language. Using latest python version's feature may make your code not work in environments of old versions.
• Install Python is boring. Linux distrubutions may have different python packages available. For e.g. installatin of python 3.7 on Centos 7 is not convenient. Though pyenv was invented for this but you still need to install pyenv firstly :).
• Python package management slightly sucks. Yes, virtualenv and pipenv are great for isolation. However, some python packages may need conflict binaries, which is not handled by either virtualenv or pipenv.
• Deployment is painful. Target system must have reuqired python version installed. This is not elegant for CI: Jenkins node has to have multiple python versions installed.

Docker Saves

• Docker can provide nearly system-level isolation.
• No need to install others apart from docker
• Dev/Pod environments keeo consistent - prefect for CI/CD.

PyQuick Helps

Though this is a super tiny python program but =>

• It gengerates a start-up python app/lib for you with Dockerfiles, SetUp Tools, Makefiles, Requirements and others ready.
• The project it generated is 100% based on Docker.
• You can immediately start VSCode to remotely code your project in container!
• Your development environment all in code. Push it to any VCS then you will be able to restore it in a few minutes by one command.

Usage

Run As Docker

The docker image is pretty small - only 40+M. So it will not take much time for you to pull it.

# pull docker image

docker pull jingweizhu/pyquick

# generate a new python app in local path: ./myproj

docker run --rm -it -u $(id -u $USER):$(id -g $USER) \
    -v ${PWD}:/tmp/local jingweizhu/pyquick \
    app /tmp/local/myproj

####  OR  ###
# generate a new python lib in local path: ./myproj

docker run --rm -it -u $(id -u $USER):$(id -g $USER) \
    -v ${PWD}:/tmp/local jingweizhu/pyquick \
    lib /tmp/local/myproj

---- OR ----

Intall From PyPi And Run It

Note: python>=3.7 required

pip install pyquick

# generate a new python app in ./myproj

pyquick app ./myproj

####  OR  ###
# generate a new python lib in ./myproj

pyquick lib ./myproj

Try Generated Project

You must have:

• docker: ">= 17.06"
• docker-compose: ">= 1.26" installed

Assume in above you input project name as mypy

Make It

cd ./myproj
make

Then check your containers

docker ps -a

A container named mypy-devenv should be running.

Run It (app only)

cd ./myproj
make run
make run ARGS='-h'

Then check your containers

docker ps -a

A container named mypy-prod should be running.

Intall To Local

If have python environment in local machine, you can install it:

cd ./myproj
make install

Build A Python Package

cd ./myproj
make dist
ls dist/*

Use VSCode To Develop Generated Project

1. Start VSCode, install Remote extention.
2. Attach to your container : myproj-devenv in VSCode
3. Open terminal. Your project folder attached to container already. Just run

   .vscode/install-vscode-extensions.sh
   

4. Reload widdow. Then python extension and other cool extensions available.

And Even More ...

Format Codes Of Your Project

cd ./myproj
make autopep8

Distribute Your Project To PyPi

1. Set up two envs:

TWINE_USERNAME=<your pypi username>
TWINE_PASSWORD=<your pypi password>

2. Run

cd ./myproj
make dist-upload

Source code

Prerequisition

• docker: ">= 17.06"
• docker-compose: ">= 1.26"

Install From Code

make install

Run

make run
make run ARGS="-h"

Dist

make dist