NUM1NEX

Codeberg, Gitlab, Tangled, anything but BitBucket.

If extremely low resource usage is required tokio+axum ain't the way to go. If your target platform is Linux with kernel version 6+ I would consider taking look at `compio` and it's HTTP library `cyper`, it has an axum adapter, so all of the axum APIs are similar.

On top of that it has `with_context` api similar to `anyhow`, that allows you to attach context and create stacktrace, when bubbling up the error.

so you can compose smaller sets into a bigger one following direction of your abstractions.

But the general pattern is - keep the errors local to your "features", rather than creating a giant global enum that does everything...

`error_set` is pretty good at it as it allows you to create `set` of errors (set per feature) and easily blend different sets together

crates.io: Rust Package Registry

We are currently exploring usage of `error_set` crate, which is similar to how Zig does it's error handling - crates.io/crates/error...

Error Handling for Large Rust Projects - Best Practice in GreptimeDB How to handle and report errors effectively in Rust applications is a common question. This blog shares our experience organizing variant types of Error in a complex system like GreptimeDB, from how a...

GreptimeDB has pretty decent error handling - greptime.com/blogs/2024-0...

Web experience in 2025

Postcard is amazing! seriously your crates is a huge source of learning and inspiration for me.

I've heard that youtube algorithm doesn't like mixing up content thematic, from the perspective of me as a viewer, I don't mind either.

Especially Apache lmao.

How efficient can zero-copy (de)serialization be? And what does it actually look like in practice?
Discover our journey implementing one of the biggest recent optimizations in Apache Iggy.

iggy.apache.org/blogs/2025/0...

#iggy #apache #asf #rust #github #oss #messaging #streaming

Could you elaborate more on what you have on mind by "dispatch" ? Do you mean actually performing some operations on the data ?

Oh wait, do you still have to toggle those ?

Finally, it was the main reason glueing me to vs code

Help us shape the future of messaging streaming! Just fill out the simple form below and share your streaming use cases with us — we’d love to hear from you!

iggy.rs/feedback

#iggy #messaging #streaming #oss

The day has finally come - our core repository has been moved under the Apache organization.
github.com/apache/iggy

#iggy #apache #asf #rust #github #oss #foss

Iggy SDK in #rust has just surpassed over 100k downloads! We're constantly improving our code, to provide the best developer experience possible, and a joy to use it when building the distributed applications :)
#iggy #sdk #asf #crates #messaging #streaming

Apache Iggy (Incubating) is built for high-performance message streaming with maximum throughput and ultra-low latency.

Explore our transparent, community-driven benchmark platform at benchmarks.iggy.rs and read more in our latest blog post: blog.iggy.rs/posts/transp...

#iggy #rust #asf #streaming

When oop purists meet haskellers.