Fixes #4: add support for no_std behind feature

[michalfita]

This PR fixes broken support for no_std by following the well recognized pattern of default std feature, which when not present turns the #![no_std] flag in lib.rs on. This method do not break unit tests or benchmarks.

Hopefully it can get merged and let nuta/kerla#70 implement ext2 filesystem using this crate.

[jackpot51]

@rexlunae could you take a look at this and merge?

[aruiz]

stumbled upon this, I'd appreciate a fix+release

[halotroop2288]

I've merged it into https://github.com/ainex-project/simple-endian-rs so we can use the master branch as an "up-to-date" version if the maintainer of this never comes back (they haven't had activity on GitHub in quite some time). I'll go ahead and merge any other important bug fixes there, cause it's important to have working dependencies.

[aruiz]

I much rather point people to crates.io than a github endpoint.

@jackpot51 forked under the Redox namespace, Jeremy and I have discussed a few times in the past about properly forking this and create a new crate that is properly maintained, I think we should just do it, we can co-maintain it and make releases from time to time

[halotroop2288]

Jeremy and I have discussed a few times in the past about properly forking this and create a new crate that is properly maintained

Good luck! I'd love to see it.

[michalfita]

What about creating an GitHub organisation and do it there? We could take a few interested parties as owners and release as maybe simply-endian.

[aruiz]

That is a great idea, the osdev-rust organization could be a great host actually, I already maintain a repo there.

osdev-simple-endian seems like a sensible namespaced fork :)

What do you think?

[michalfita]

Everything that works. Let's make sure the crates have a bus set up (does that initiative still works?).

[mkroening]

FYI: I have just published rust-osdev/endian-num with a focus on following core::num conventions. :)