feat: L2Resolver state variables and constructor w/unit tests

[0xItadaki]

🤖 Linear

Closes EFI-425
Closes EFI-434

[linear]

EFI-425 constructor and variables

State Variables:

• mapping(bytes32 _node => mapping(string _key => bytes _data)) public records;
  • Stores arbitrary data for each ENS node.
  • For domain-to-chain-identifier resolution, the key will be a pre-defined string (e.g., CHAIN_IDENTIFIER_EIP7930_KEY) and data will be the raw EIP-7930 chain identifier bytes (formatted with AddressLength = 0).
  • For EIP-7930-chain-identifier-to-domain resolution (reverse lookup), the REVERSE_LOOKUP_NODE is used. The key is constructed by concatenating a prefix (e.g., CHAIN_IDENTIFIER_EIP7930_KEY) with the hexadecimal string representation of the keccak256 hash of the EIP-7930 chain identifier bytes. The data stored is the ABI-encoded human-readable ENS name string (e.g., abi.encode("optimism.l2.eth")).

Constants:

• bytes32 constant REVERSE_LOOKUP_NODE = bytes32(keccak256("reverse.chain.id.eip7930"));
  • A dedicated node hash for storing reverse lookup entries within this contract's records mapping.
• string constant CHAIN_IDENTIFIER_EIP7930_KEY = "chain.id.eip7930";
  • The key used in records to store the EIP-7930 chain identifier bytes for a forward resolution.

EFI-434 constructor

[0xItadaki]

@0xAustrian what do you think about the visibility of these variables? I went with public so I could unit test them. I kept _REVERSE_LOOKUP_NODE since it will be used in internal logic

[0xAustrian]

I would set everything on public tbh, even REVERSE_LOOKUP_NODE

[0xItadaki]

do we usually set the author?

[0xAustrian]

Yes sir

[0xAustrian]

Empty space here

[0xAustrian]

Let's use a fixed .sol version.

[0xItadaki]

why is that?

[0xAustrian]

I mean it's no big deal but a good practice to use the exact same version everywhere.

[0xAustrian]

This is so the CI does not fail, right? In that case I think we can temporarily comment the integration tests section in the workflow (sorry, forgot to mention that. It would have been a good addition to the PR#1).

[0xAustrian]

Good job!