Closed lykahb closed 8 months ago
Thanks for the review @infinisil. I could see how Base16ByteString could be misinterpreted that way. Instead of PrintableByteString
I chose a name PrettyHexByteString
instead - it both shows that its main purpose is pretty-printing and that it displays it as hex. Let me know what you think.
The documentation for the COSE newtypes points to the rfc8152. I am not deeply familiar how COSE should work, so can you take a look at that?
The haddock has full coverage with no warnings.
This is based on https://github.com/tweag/webauthn/pull/144
Compared to that PR, this branch:
It seems that @infinisil's concerns about the COSE signature and the spec are related to the choice of algorihtm, and are orthogonal to the refactoring that introduces newtypes
CoseMessage
andCoseSignature
.