feat(consensus) implement RLP for Account information

[DoTheBestToGetTheBest]

Closes #780

[prestwich]

Based on the reth struct here, we should be able to derive this implementation. However, we will need to re-order the fields. That seems fine to me

while we're here, I would also like to see a convenience function for trie_hash (similar to in Encodable2718

pub fn trie_hash_slow(&self) -> B256 {
    let mut buf = vec![];
    self.encode(&mut buf);
    keccak256(buf)
}

bonus points for memoizing the hash in a OnceLock

@Rjected @mattsse do you have any RLP encoding/trie hashing test vectors handy?

[mattsse]

not sure, I don't think we use this type in reth because we don't store the account's storage root

maybe @rkrasiuk knows if we have some test vectors

but I wonder if we can use derive here?

[prestwich]

not sure, I don't think we use this type in reth because we don't store the account's storage root

hmmm, this type is committed in the state MPT right? or am i misunderstanding?

[DoTheBestToGetTheBest]

Based on the reth struct here, we should be able to derive this implementation. However, we will need to re-order the fields. That seems fine to me

while we're here, I would also like to see a convenience function for trie_hash (similar to in Encodable2718

pub fn trie_hash_slow(&self) -> B256 {
    let mut buf = vec![];
    self.encode(&mut buf);
    keccak256(buf)
}

bonus points for memoizing the hash in a OnceLock

@Rjected @mattsse do you have any RLP encoding/trie hashing test vectors handy?

Thank you very much for your review i added the trie_hash and trie_hash_slow methods

[DoTheBestToGetTheBest]

but I wonder if we can use derive here?

should i directly derive ?

[prestwich]

yes, please derive and reorder the fields to match reth as linked above

[DoTheBestToGetTheBest]

yes, please derive and reorder the fields to match reth as linked above

Thank you very much guys for the review of the PR.

I reorder the field similar to reth struct and i derived instead of manual impl

[prestwich]

thank you. last few things:

• after some thought, I think we're not clear on our memoization strategy yet. So let's go ahead and remove the trie_hash fn, and leave trie_hash_slow. This will give us some time to think about how we want to do this across the code base.

• I would like a couple tests, because this is consensus critical code, and used in reth consensus at some point. You don't have to provide tests right now. I can open a follow up issue.

[prestwich]

oh and for no_std support in CI, need to add a

#[cfg(not(feature = "std"))]
use alloc::{vec, vec::Vec};

[DoTheBestToGetTheBest]

thank you. last few things:

Thank you very much for you review & your patience !

• after some thought, I think we're not clear on our memoization strategy yet. So let's go ahead and remove the trie_hash fn, and leave trie_hash_slow. This will give us some time to think about how we want to do this across the code base.

No problem !

• I would like a couple tests, because this is consensus critical code, and used in reth consensus at some point. You don't have to provide tests right now. I can open a follow up issue.

Thank you, when you open issue about it if it's possible to link any example of test like this i'd like to take that new issue :)

[prestwich]

nit:
we usually write this using the conveniencealloy_primitives::keccak256 instead of invoking Keccak256::new() and update

[DoTheBestToGetTheBest]

nit: we usually write this using the conveniencealloy_primitives::keccak256 instead of invoking Keccak256::new() and update

Thank you very much, i changed !

[prestwich]

Thank you, when you open issue about it if it's possible to link any example of test like this i'd like to take that new issue :)

this is what i was asking mattsse and rjected about earlier :) I am not sure where to link for testvectors for account encoding

[DoTheBestToGetTheBest]

Thank you, when you open issue about it if it's possible to link any example of test like this i'd like to take that new issue :)

this is what i was asking mattsse and rjected about earlier :) I am not sure where to link for testvectors for account encoding

Yep gotcha, ty very much for all support on this !