Make the articles featured in Docker Weekly available via JSON and eventually, via a hosted service that can be consumed as an API.
- Run
docker run --rm -p 8080:80 infracoder/docker-weekly-api
- Go to http://localhost:8080/
The motivation for building this was that the Docker Weekly (at time of creation, April 2017) was not being updated to include the issues from 2017. This means that new subscribers had no way of seeing previous issues.
All articles in each Docker Weekly issue from 2016 and 2017 have been parsed and put into the data directory.
Just git clone
and inspect the data
directory for the issue data.
While Docker has gone through many rapid changes since its inception, many of the articles captured in the Docker weekly articles arwe timeless as they focus on best practices and lessons learnt using Docker in production.
I would love help turning this into something more beneficial to the community. Pull requests are welcome.
See the TODO.md
file for the most urgent needs.
Raise an issue if you want to suggest new, different or complimentary data to what's currently available.
Currently, you'll need a Python environment with version 3.6.x
installed, the easiest way of which to do this is, of
course, to run it in a Docker container.
Then in the root of the project, execute export PYTHONPATH=$PWD
so Python will look in the root directory for modules.
This project is in its infancy and exists as Open Source more for transparency reasons than anything else. As a result, it's not really yet ready for contributors, unless they're ok with the complete lack of documentation about how to get started.
But, this will change once the service is online, working and providing value. At that point, I'll switch gears and focus on how others can take this shell of an idea and turn into something great that hopefully, provides real value to the Docker community.
Running the bin/get_issue.py
script may result in replacing complete data with incomplete data!
As not all Docker Weekly issues were able to be automatically parsed, manual work copy-and-paste work was done to finish the population of the data/issues
files.