Daniel McCarney

I don't believe this issue is relevant with tip-of-main. If I'm mistaken and someone can reproduce with tip, or the 0.14.0 release, please comment and we can revisit.

This looks easy to fix, but I'm not sure it's worth breaking semver for. In the primary rustls repo we have a `next-major-release` label for this. I've created the same...

_(Was this meant to be filed on [rustls/rustls-ffi](https://github.com/rustls/rustls-ffi) instead of this repo?)_

@jsha I started working on the "Stage I" plan and have a prototype branch that I think gets us most of the way there: [cpu-choose-your-own-crypto-adventure](https://github.com/rustls/rustls-ffi/compare/main...cpu:rustls-ffi:cpu-choose-your-own-crypto-adventure). * `make` is our status-quo:...

Thanks for the feedback! > Right now the only thing you can do with a *rustls_supported_ciphersuite is get its IANA registered number, or its name. I think we could cut...

Here's [a checkpoint branch](https://github.com/cpu/rustls-ffi/tree/cpu-choose-your-own-crypto-adventure-wip-checkpoint) that's a bit further along (but still missing lots of polish). * The connection type now lets you return the suite ID or name of the...

> To match the Rust API I think we need to expose a type for Arc and a function that parses PEM. That function should collapse some of the complications...

I was able to find time to come back around at this. PTAL: https://github.com/rustls/rustls-ffi/pull/441

I believe the needed API surface has been added for client certificates and CRL based revocation. If there are specific needs outstanding I think it would be best to follow-up...

I had been idly thinking about doing a pass to remove type annotations where not necessary and then I found there was already this issue about the idea. I wonder...