Dirkjan Ochtman

Ah, that's a a good bit more complex than what we have now. So if I understand correctly: * There might be two `postalInfo`s, not just one (as we have...

This sounds reasonable to me. This might be an oversight from when we migrated to the generically-typed `RData` setup, or it might be necessary in some contexts to take out...

Strong 👍 from me!

I'd be open to a PR that introduces the use of cargo-hack similar to what we have in the rustls repo. Might want a separate job for it since it...

> I keep hesitating on the removal of OpenSSL because of some of the compliance it has that we're still waiting on for Rustle. I think we should just clean...

Yes, Rustls from 0.22 onwards supports the aws-lc-rs crypto provider which has a FIPS mode available (on Linux only).

Thanks for the report! Could you send your test case as a PR adding a unit test in `crates/proto/src/rr/serialize/encoder.rs`? Bonus points if you can also come up with a fix,...

> Agree it is not actionable locally and quite verbose, especially when you have only few direct dependencies and that's very noisy. (..) Putting in behind `--verbose` would reduce the...

> > If anything, I would be happy if you showed Unchanged semver-incompatible bumps for my workspace's direct dependencies, but it currently doesn't -- so that also feels inconsistent. >...

> With `cargo upgrade`, didn't you push back on a lot of our output unrelated to the actual upgrade, having us put it behind `--verbose`? Showing unchanged direct dependencies (that...