You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Both styles are a bit different since go-ipa expects an opaque stream of bytes and does the deserialization. Here, in go-verkle, we have the serialized proof already in parts, which probably explains why we have some custom code.
In any case, we must rely on go-ipa to do any deserialization to know this is done safely. If that means changing go-ipa APIs since we need to pass these "parts", I think that's fine. But let's try to rely on go-ipa for any security issues since it has (and will continue to have) more tests (and audits) to be sure these APIs are safe.
The text was updated successfully, but these errors were encountered:
Yeah this is the old API, it has some issues. We would have to revisit it before we move back to this system. I don't oppose it, but let's keep things as they are now for kaustinen. It's much easier to debug.
Currently, we have some custom code to deserialize a
Multiproof
. In go-ipa, there's a specific API to do this deserialization.Both styles are a bit different since go-ipa expects an opaque stream of bytes and does the deserialization. Here, in go-verkle, we have the serialized proof already in parts, which probably explains why we have some custom code.
In any case, we must rely on go-ipa to do any deserialization to know this is done safely. If that means changing go-ipa APIs since we need to pass these "parts", I think that's fine. But let's try to rely on go-ipa for any security issues since it has (and will continue to have) more tests (and audits) to be sure these APIs are safe.
The text was updated successfully, but these errors were encountered: