speer VS StreamSaver.js

So first I generate a token using the jsonwebtoken crate. This accepts a struct that is serde serializable and deserializeable. After that I construct the Cookie and send it as the response to the request for the '/login' route. You can see my struct here (I named it JWT, but that is not the best name ever, this is the data included in the JWT token): https://github.com/horvbalint/speer/blob/main/backend/src/jwt.rs And then the login route here (line 88): https://github.com/horvbalint/speer/blob/main/backend/src/routes.rs

I just finished rewriting the backend of my side project from Node.js to Rust (actix-web). This is my first 'big' Rust project and I am sure that there are errors/bad practices in my code, so I am looking for some more experienced Rust developers who can review my code and give me feedback. I marked some places in the code, where I am sure that the code can be improved a lot. Here is the repository https://github.com/horvbalint/speer/tree/WIP---backend-Rust-rewrite. The README contains (hopefully) every information needed to get started.

You can try it out here and read more about it and check the repository out here

This! It is frustrating that web apps like draw.io can't save to file directly using Firefox. It gets old having to save a new file every time you make a change.

Beyond that, the File System API would also allow for streamed downloads directly from the web-app. This would allow for web-apps to generate large download files without having to store a copy on a database (external server or IndexedDB), which is a privacy improvement. The only way to do this now is by using convoluted techniques to essentially do a MITM to a fake endpoint [1].

[1] https://github.com/jimmywarting/StreamSaver.js

Yuck, that sounds annoying. I'm on Linux and generally use Firefox, so I haven't run into that yet. I'm using https://github.com/jimmywarting/StreamSaver.js for saving data to disk, and it seems that a similar issue may be present for that project as well.

If it's larger, it's complicated... https://github.com/jimmywarting/StreamSaver.js generally works, but is not 100% reliable. https://github.com/jimmywarting/native-file-system-adapter is supposed to be like a newer version of it, but I struggled to get it working. You can also use https://web.dev/file-system-access/ but it's only supported in Chrome.

The hacky new way. There is a library which lets other browsers stream data to a file. Problem is, it's a bit unreliable. In my testing, I found it sometimes doesn't work in Firefox (reload usually fixes it) and it didn't work at all in mobile Chrome (not sure why yet).

Thank you for the kind words! Firefox is not fully supported, because of two things: - no support for WritableStream (used for downloading the files). Read more here - not supporting system audio capture at all Non of this is breaking though, and file downloading still works, but it might be a little slower.