Deploy python package with private github repo
Kiran Koduru • Jul 3, 2017 • 1 minutes to read
If you are someone who hosts your python package in a private github repository for work or someone who plans to develop your package privately before it is open sourced, I have answers to your questions on how to deploy your project without hosting your own private PyPI service.
First, I hope you have followed the guide on how to distribute your package. If not, this might come in handy. At most you have a setup.py file in place for your package. Next you can follow these steps for a elaborate walk through.
1. Create github user
Create a github user with access to this one private python repository. This is in case your access token in Step 2 is exploited, this step will mitigate any risk to your access token.
2. Create a github access token
Login as user from Step 1 and create a new access token with Full control of private repositories as scope. This will allow to read, write and delete your private repo. Hence creating a new github user will minimize the risk to just this repo if your access token ever gets compromised.
3. Tag your git repository
Next, tag and release your private github repository. Either use the github UI to tag your release or if you would like to use the command line
git tag 0.1.0 git push origin --tags
This steps allows you to make rolling deploys in case you don’t want to deploy the updated projects all at once or have different versions in different projects.
4. Add to requirements.txt
Finally update your requirements.txt file in the python project that depends on this private project and let pip take care of the rest of it.
git+https://{access_token}:x-oauth-basic@github.com/{username}/{project}.git@0.1.0
What I mean is if your deployment practices follows installing packages with
pip install -r requirements.txt
then your project will update the package version with each deploy. Make sure you’ve replaced {access_token}, {username} and {project} appropriately for the command above.
I am writing a book!
While I do appreciate you reading my blog posts, I would like to draw your attention to another project of mine. I have slowly begun to write a book on how to build web scrapers with python. I go over topics on how to start with scrapy and end with building large scale automated scraping systems.
If you are looking to build web scrapers at scale or just receiving more anecdotes on python then please signup to the email list below.