[cryptography] Refactor Scheme
#991
Conversation
cryptography/src/lib.rs
Outdated
| message: &[u8], | ||
| public_key: &Self::PublicKey, | ||
| signature: &Self::Signature, | ||
| public_key: &K::PublicKey, |
There was a problem hiding this comment.
I think we may be able to get away with impl PublicKey and impl Signature?
There was a problem hiding this comment.
I suppose they should be fixed for a single instance of a BatchVerfier (so maybe we should make both generic)?
There was a problem hiding this comment.
I'm not sure I follow. We can't put impl PublicKey / impl Signature because a BatchVerifier is parameterized on a PublicKey type, so it can only verify using its PublicKey type and associated Signature.
Scheme
Codecov ReportAttention: Patch coverage is
@@ Coverage Diff @@
## main #991 +/- ##
==========================================
- Coverage 91.18% 91.09% -0.10%
==========================================
Files 191 191
Lines 54967 54836 -131
==========================================
- Hits 50124 49954 -170
- Misses 4843 4882 +39
... and 1 file with indirect coverage changes Continue to review full report in Codecov by Sentry.
🚀 New features to boost your workflow:
|
Resolves: #755
Resolves: #1049
Unblocks: #432
This PR updates the cryptography traits defined in
cryptography/src/lib.rs. Notably, it removesSpecificationandScheme. The new crypto design has several interlocking traits:Signeris capable of producing a signature over a messageVerifieris capable of verifying a signature over a messagePrivateKeyis aSignerthat is serializable and had aPublicKeyPublicKeyis aVerifierthat is serializableThe vast majority of the diff is updating downstream code. There are also many generic types
C: Scheme(Schemein the old code is most analogous toPrivateKeyin the new code) that have been turned intoC: PublicKeybecause they don't rely on private key capabilities.