-
Notifications
You must be signed in to change notification settings - Fork 141
Support for more backends #1320
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
Hi. This is a good start for some limitations of pure file storage protocols when it comes to (Web) sync: |
@raucao thanks. Any thoughts/talk on s3 and github? |
I would say S3 has almost exactly the same drawbacks as WebDAV, except that it's even more difficult to get credentials as a normal end user. Not sure how you would imagine GitHub to be used as personal data storage exactly. Care to elaborate? |
I was trying Diffuse, and it seems to be able to connect to s3 ok. You need to set CORS on the bucket: There is a question of what a "normal end user" is here. Which I am sure has been discussed. I would hope a user who is empowered to bring their own storage could be technical or not.
A git repo stores the state. You can probably issue a api key for your repo in github. Then use that key to read and write the contents of local state. Communicate over github's restapi to issue reads and updates. |
I am looking for a front end lib that can 2 way sync to various filestore backends, based on the users choice. Without the need for the user to run their own service.
I do see that Solid is in the works, and thats great.
I'm imagining something like Rclone that supports a bunch of backends...
Open ones like: SFTP, SMB, WebDAV, nextCloud
and hosted ones like: s3, backblaze, github, etc.
while reading and writing to these may not be too hard, 2-way sync is another problem and I have not found a generic solution for those.
Of course this is no simple task. Each backend is its own can of worms.
But since I have not found discussions here on other backends, I wonder if there are high level limitations that would make these difficult. Perhaps oauth, or metadata related? If I were to take a crack at implementing one, are there pitfalls I should be aware of?
The text was updated successfully, but these errors were encountered: