search

hagsteel / swampdragon

swampdragon
Other
557 stars 73 forks source link

0.5 and going forward #161

Open hagsteel opened 9 years ago

viaregio commented 9 years ago

I'm using SwampDragon in my django project and it works great. I am a little bit worried that the lost of focus (not django only anymore) may slow down the development or that some features getting dropped (ModelRouter for instance).

hagsteel commented 9 years ago

No features should be removed, but as it is right now it's not overly open to other frameworks. The routers are core to SD, so they will not be dropped. This doesn't mean that routers couldn't work without Django (there is no dependency on Django in the actual route_handler.py).

The support for different serialisers should be added, not just DRF as there are other serialisers out there, so the entire handling of serialisers could be reviewed.

There is also the dependency of a library that is no longer actively maintained, this is a big concern, so the dependencies needs to be revised as well (I am exploring some options as I'm writing this).

The general usage of SD seem to be designed well enough since most people get it, so the aim is not to change anything major there.

The self publish model needs revising.

There has been great support from other users who are helping out by answering questions, creating PRs etc. and I would love to have more people as actual contributors.

There is a need for better testing (and more tests!)

update: I re-read the note in the README and I can see how it came off as "Django needs to be removed" so I updated the README to avoid missunderstandings

denizs commented 8 years ago

Hey Jonas! I am interested to participate in taking swampdragon to the next level! Let me know if there are any new plans and tasks in the pipeline. Best, D

hagsteel commented 8 years ago

@denizs I think the original plan might be a bit dated by now.

The removal of Tornado was one of the ambitious steps I wanted to take, but that would've meant removing Python 2.x support (using asyncio). It's a shame because asyncio could've meant many of the dependencies could've been ditched.

I think the main things to do are reviewing and merging/dismissing all the PRs.

I would actually recommend you fork this and try to keep it going, and I can put a link in the README to your repo.