dnf VS crates.io

https://github.com/rpm-software-management/dnf/pull/1879

However its o my going to work on rhel 10 or anything that has very very up to date DNF version.

I'm not sure, I'm not a dnf developer. It was implemented in C++, so it would probably have to be done differently in version 4. I belive dnf development is focused on version 5 right now, so adding features to version 4 is probably not a high priority. It would be nice to see it in version 4 so it could be added to current versions of RHEL and Fedora, but time is a finite resource. Perhaps you could implement and send a pull request, or at least file a bug to discuss it upstream?

dnf uses python only for command line functionality. Its core part, libdnf, is written in C++. However yes, dnf in my experience is still significantly slower than apt and pacman.

Lots of people use python apps and not know about it. I don't get where the idea that it's complicated comes from. You run them the same way as everything else. For example if you use anything redhat-based you're running a python app (https://github.com/rpm-software-management/dnf). If you're running any desktop environment, you're likely using at least one python app or something with python scripting embedded.

dnf: https://github.com/rpm-software-management/dnf/pull/1602

dnf for Fedora and Redhat Linux.

Hey I published my first crate on crates.io and it tells me that 45 people downloaded it. That's hard to believe. Is it a known fact that the download numbers are higher because of any reasons (for example bots even though I don't see how that makes sense) or should I just be happy that people are using it? Im not sure how 45 people should even know about this crate. Thanks!

Haven't done a release yet to crates.io, but will consider it at some point once I have a few more things nailed down.

Having worked with bazel (the public version of Google's build system). You can make any build hermetic, but sometimes it's a case of how much effort you have to put in. And yes, crates.io dependencies which use build.rs can be (though not always are) a real make to make hermetic.

Developing the compiler? Maintaining crates.io? Helping organizing conferences? Helping users on the official forums? ...? All things have value, but there is no one-fits-all answer how to start.

Standardized documentation for all libraries in ecosystem. You can just open almost any decent crate in crates.io and you would know where to look for examples and such.

You can also configure where the dependency is pulled from. By default, it gets pulled from crates.io central registry. You can also pull specific commit from specific repository, or even from local folder.

Here's something I came up with quickly. https://github.com/TukanoiddExperiments/variant_common_data. It's not perfect, can be improved a lot (feedback & PRs always welcome! I intend to use this crate myself in the future, since I've had to deal with this issue quite a few times). For now, it's gonna stay in my sandbox workspace before it's ready for 0.1.0 crates.io release. Also, I'm very bad at naming, so ideas on that front are also welcome! I already have some ideas regarding additional features and error handling improvements, but I'd like to hear other peoples thoughts on what I should use/switch to for better compile times, I don't write proc macros very often, so my knowledge on the proc macro crate ecosystem/idiomatic way to parse things/etc might be outdated.

I've been looking for such a crate on crates.io but all I could find was the arrayfire and Rustacuda crates. Arrayfire seems out of the question for me because I have a function that is meant to be executed on all threads of the GPU, each execution of the function on a GPU thread returns a simple struct of 3 floats. Rustacuda is kind of what I'm looking for but unfortunately it only supports calling C Cuda external functions and does not support calling Rust functions you created.

I've finally released my first crate to crates.io.

Published to crates.io here https://crates.io/crates/safe_decimal but please - it's not a crate ready to be used in production.