Future of this project?

[marteinn]

Hi there! You might have noticed me submitting two PR:s the last couple of weeks, I'm curious about the future of this project and how it can be more actively managed.

The reason is that I maintain a library called https://github.com/Frojd/django-react-templatetags and are looking into replacing our current SSR service with Hypernova. I did some research and found this library to be the most suitable, primarily because the code is easy to read, it does whats its supposed to without any major 3-party packages. (Yelp also has an interesting library but it is too reliant on Pyramid.)

But before adopting this library a few things needs to be done:

• Make it fully python 3 compatible (right now this causes errors when using with the dev plugin)
• Drop python 2 support (this can be discussed but with the python 2 EOL this a good chance to simplify management)

(I have a fork where you can see some of my fixes/implementations: https://github.com/marteinn/hypernova-python/). You can also see the hypernova integration in DRTT here: https://github.com/Frojd/django-react-templatetags/blob/develop/django_react_templatetags/ssr/hypernova.py

So my question is how - can I contribute without being a nuisance? :) (I know how open source work and just because you published a library does not mean just have any obligations)

My thoughts are this:

• I submit PR:s like I've done before and you publish updates
• Or you let me onboard as a maintainer or transfer project
• Or I continue with my fork

Let me know what you think and thanks for your work so far with hypernova-python!
/ Martin

PS: Sorry for the long issue!

[ornj]

Hi @marteinn, not a nuisance at all. I haven't been actively maintaining this project as it not relevant to my current life, but I wouldn't say I've abandoned it.

I very much appreciate the contributions and they nudge this project a little higher in my mental queue. If people are using it I'm more interested in actively maintaining it. My short list of tasks has been:

• Drop Python 2 support (at the time I needed it)
• Improving test coverage (Trade Travis CI for Github Actions, write some e2e tests)
• Improve the API (this was implemented with an eye towards https://github.com/airbnb/hypernova-node and https://github.com/airbnb/hypernova-ruby, there are things those libraries do that I would like to replicate such as expanding the API, batch rendering, and being able to drop in other http libraries)

At the very least you should be a contributor. The only reason I hesitate making you a maintainer is that this is our first introduction and the project doesn't have clear guidelines. For the time being I say continue your fork and I'll be diligent about reviewing any merge requests.

[marteinn]

Thanks for the quick response @ornj!

Awesome to hear that you still have a interest in this project, looking at your shortlist (which fits well with my near future goals) I'll start with creating a PR on the theme of dropping Python 2 support and making sure Python 3.5+ works as expected and go from there, sounds good?

PS: Batch rendering sounds very interesting seeing how we render components in DRTT :)

[ornj]

I made an effort to get some code style / formatting standards in place as well moving ci from travis to github actions. As a part of that I removed six and focused on testing Python 3.6+. Submit PRs for any other refactoring that might need to happen to drop py2. I haven't generated any release as "dropped python 2 support" doesn't seem like much of a release to me.

Beyond that I don't expect to be too active here besides reviewing potential contributions. At least not at present. If you wanted to give me a sense of direction you think the project should go that would help me identify if we seem to have alignment on expectations / philosophy and if making you a maintainer would be a good fit.

[marteinn]

Awesome, looks great! I would actually recommend that we do a new release to solve the devplugin bug (you currently cannot use it in py3).

Honestly I am very early in my hypernova adoption, I will probably have my clarity once my company starts replacing our internal SSR tool with hypernova and of course feedback coming from users of DRTT.

But vision wise:

• Keep the library with a light footprint with as few deps as possible so it can be something that can be integrated in other libraries with ease. You have already done a good job on this.
• Not breaking api's, I shouldn't have to say this but.. :)
• Examples and improved docs to get more usage and traction
• Some sort of development toolkit in a separate repo
• I think end to end tests are a good idea, should be pretty straightforward to do in github actions with docker
• A automated release process for the python library would be a good idea as well
• Adopt a versioning schema (semver?) and updating pypi classifiers
• Reading through the hypernova spec and implement more features :)

I have a long term and goal when it comes to django and that is to do something like https://github.com/reactjs/react-rails. Its a long way to go but I currently have DRTT for templating and a react based Wagtail boilerplate (https://github.com/Frojd/Wagtail-Pipit). I think hypernova and hypernova-python fits well in that vision.

[marteinn]

Hi @ornj, just wanted to say that I released an updated version of https://github.com/Frojd/django-react-templatetags a couple of days back and it now includes your Hypernova library, thought you should know :)

[ornj]

Adding you as a collaborator as you obviously have on going interest and I can't prioritize this at the moment. Let me know if you need anything else.

[marteinn]

@ornj Awesome, I won't disappoint :)