Georg Brandl

Yes, I'm fine with it if it's explicitly requested. This command line switch would also be exposed as a flag in the API, I assume?

Sorry, no, I think you're good. We just need some cycles to get to it. I added the 2.12 milestone now to ensure you're not forgotten.

Looking at the CI, a rebase should fix the error there.

You can also merge with master, but for smaller PRs it's usually preferable to rebase on top of master, since it avoids the commit history getting a tangled mess :)...

Great! I was wondering about tests, but not sure how easily this can be tested. If you can think about something, give it a try :)

Please give me time to review this, currently life is in the way...

Another valid use case that came up [on Reddit](https://old.reddit.com/r/rust/comments/ctayew/rust_crypto_developers_please_stop_the_yanking/): allowing people to do `git bisect`.

I volunteer for that.

Sounds good to me. There is extensive material on the Python side on how these protocols work, and we should mainly state what we support and what the Rust/PyO3-specific snags...

well, C++ enums are a fair bit different from Rust enums.