Closed IngridPuppet closed 10 months ago
Is this really necessary? Cargo flags this as unused.
This will panic if the incoming slice is not of the right length. Maybe you'd want to use the same logic as with from_secret_key
.
There is a nicer API for this line: sk.verifying_key()
.
Vec::new()
looks like a sentinel value, which does not align with Rust philosophy. Why not return an Option?
L83: Why not emit a meaningful error instead of panicking? L86: I noticed there is an Error::SignatureError variant. Does it not belong in here?
This should test instead for keypair.verify
. Isn't it?
ChatGPT was definitely not great here. The key_exchange
method should be used at least twice for both parties, in lieu of diffie_hellman
, which is the goal IMO.
@francis-pouatcha : Did not get your point here. Can you explain?
Maybe substitute an assertion for the print?
All fixed except this: https://github.com/ADORSYS-GIS/didcomm-mediator-rs/issues/15#issuecomment-1701190321
Did not get the point.
Please review and move to done.
Below are point-to-point questions and suggestions.