ektrah

Fwiw, I think it's totally the right way to do (passing keys around as `KeyPair` instances rather than byte arrays & storing private keys in *sodium_malloc*'d memory). The "problem" I'm...

Nice work 😃 The artifact (libsodium.so) also needs to be added to [libsodium.pkgproj](https://github.com/jedisct1/libsodium/blob/master/packaging/dotnet-core/libsodium.pkgproj) so it gets included in the NuGet package. /cc @SeanMollet @enclave-alistair

libsodium.pkgproj looks good to me. While you're at it, you could also increment the version number in the same file. Now someone needs to go ahead and test if the...

> Happy to bump the version - to 1.0.19.0? That would be 1.0.18.3 (the third revision of the NuGet package containing pre-compiled binaries of libsodium 1.0.18).

@MichalPetryka, [.NET doesn’t run on s390x](https://github.com/dotnet/core/blob/main/release-notes/7.0/supported-os.md). (Note that this particular PR is about the packaging of libsodium for .NET, not libsodium in general.)

@sigaloid I'm mostly concerned about the API design. It seems that a lot (most?) projects on GitHub that reference NSec contain some variant of the following code: ```csharp class MyKeyPair...

Yes, that would be a good enhancement. What specification should NSec implement to provide that? I was thinking of [RFC 5958, Section 3](https://tools.ietf.org/html/rfc5958#section-3) (EncryptedPrivateKeyInfo), but that covers only asymmetric keys.

Thanks for sharing your thoughts on this topic. I agree that the state of libsodium bindings for .NET is not very satisfying: libsodium-net is dead, the development of libsodium-core seems...

Nice work. I’d be interested in how you’re planning to make libsodium misuse resistant (or if you’re planning to just drop everything that isn’t?). Overall, I think it’s clear that...

In short: Every time you encrypt, you *must* use a nonce that has not been used before (with the same key). When you decrypt, you *must* decrypt using exactly the...