NEP-message-signing #213
NEP-message-signing #213
Conversation
| wallet sign message <message> | ||
| </pre> | ||
|
|
||
| A conforming implementation MUST: |
There was a problem hiding this comment.
The standard is written in a way that ensures there is a single implementation. NEPs shouldn't specify CLI commands.
There was a problem hiding this comment.
thank you Roman. I'll rewrite it.
There was a problem hiding this comment.
I'd like to add that it be described in a language agnostic way. I see a lot of C# code in there
| That is: | ||
|
|
||
| * 4 bytes: `networkMagic` in little-endian. | ||
| * Followed by 32 bytes: the serialized `hash` (`UInt256`). |
There was a problem hiding this comment.
Not sure why UInt256 got involved here.
Can't we just say something along the lines of
The data to be signed using secp256r1 is a byte array consisting of; a + b where
ais the networkMagic encoded as a 32-bit little-endian unsigned integerbis thesha256(sha256(payload))digest wherepayloadis formatted as described in<section xyz>
All these C# code references from the neo project make no sense for other languages and should be irrelevant for a specification.
Instead, I would suggest to add a section with test vectors so implementers can verify their implementations.
There was a problem hiding this comment.
thank you @ixje I'm rewriting the proposal. It will be uploaded next weak.
Add message singing proposal