Jonatan Männchen

Blocked by https://github.com/erlef/setup-beam/issues/199

Shouldn’t `accept` use the `hex` type instead of `otp`? https://github.com/package-url/purl-spec/blob/master/PURL-TYPES.rst#hex https://github.com/erlef/security-wg/blob/master/docs/specs/otp_purl_type.md#relation-to-hex-purl-type

Yes, exactly that. This was just something like I stumbled upon when trying to refactor some code. I have a working solution right now, so this is not at all...

@kipcole9 That is awesome. :heart: So the missing piece if I understand correctly is that we have to identify if there are compatible units in the dividend and the divisor....

Thinking about it a bit more, we probably have to take a less naive approach to solve those. GNU Units does what we're looking for: ```sh units "8meter/second * 300second...

That is correct. We can’t differentiate the nodes. If you have a good idea how to do that, please open a PR. Otherwise you‘ll have to implement a custom strategy...

> I think there shouldn't be a default for client_secret_jwt or private_key_jwt. @paulswartz I tend to agree. I think I remember some certification issue though without a default. I’ll rerun...

@paulswartz Are we able to sign the token using the algoithms provided? On a first glance, they all look like they are asymmetric… I’ll have a look myself in the...

@paulswartz I see the reason why it was in: **Certification Setup**  **Test Command:** ```bash mix run_certification \ --profile test \ --alias test \ --version v3.2.0 \ --auto-open \ --auto-screenshot...

I've started implementing a standalone http message signature lib that can later be used here: https://github.com/maennchen/http-message-signatures