As many of you here, we have been affected by the deprecation of Realm services. As we manage several production applications, we had to find a way not to disrupt the service offered to our customers.
Our decision was to rewrite the core using Node.js and TypeScript, keeping the same folders structure and configuration format as the deprecated services: we needed a solution that could easily be swapped in without major changes to our existing projects.
Our rewrite is not complete yet, as not every Realm feature has been ported. The focus has mostly been on authentication, rules, functions, triggers, services, http endpoints.
Eventually we decided we could try the open source route sharing our work with the community, in the hope of helping others that faced the same problem and maybe encouraging other people to take part in this rewrite of the deprecated services.
Before sharing our work we would also like to know the opinion of MongoDB.
Thanks for sharing! I’m a Lead Product Manager here at MongoDB on the Developer Experience team, so I wanted to at least reach out. I would personally encourage our community to share their solutions and feel free to promote them on our forums.
If you have specific concerns that are giving you pause before releasing the solution, I’d be happy to answer any questions you may have - or get the answers if I can’t answer them myself.
Feel free to ask here, shoot me a DM or send me an email.
I’ve also worked on this project: a Fastify based application that works almost exactly like the Realm platform (minus the sync and other features that we are not currently using); we also kept the structure/configuration of consumer projects the same as if used with Realm, to avoid significant changes in our production applications.
We are almost ready to make the repository public and completely open source, as we’ve completed the first successful migration of one of our existing apps.
Since our project is basically a smaller rewrite of your original platform, we would like to know if we can share it without limitations and encourage other people to try it, give their feedback and contribute if they want. If that’s ok we’ll soon share the link to the repo with a demo project.
Hey @Gianluca_Spada, if the Stackhouse team has a solution for the community they want to share as an open source solution I see no reason why you shouldn’t share it.
We shared additional feedback on the decision to EOL device sync (I see the Stackhouse team recently commented there as well) on Atlas Device Sync: End-of-Life and Deprecation, and though it may not be the most popular product decision we’ve made, if the community has a solution we want to encourage and foster the spirit of ongoing open source development.
I look forward to seeing what your team has come up with, as I’m sure many Atlas Device Sync users are as well!
Thanks,@alexbevi! We really appreciate your support and encouragement.
As mentioned, we’ve been working on a lightweight alternative inspired by the original platform, and we’re excited to finally share it with the community!
GitHub Repository: 
NPM Package: