This PR adds a new column sourcify_matches.metadata. The contract's metadata is going to be stored in this new field. This new field is going to be used by SourcifyDatabaseService as metadata's source of truth instead of RepositoryV2
This PR also add a new check in tests: keccak256(repository metadata) === keccak256(database metadata)
Metadata is always stored in database even for cases in which contracts don't have a metadata field in the bytecode (just like we do on repository)
Doubts:
We are going to rebuild production's database from scratch so we are not going to have problems while applying ALTER TABLE sourcify_matches ADD COLUMN metadata json NOT NULL. How do we want to handle this case on staging? We are going to rebuild also staging database from scratch
sourcify_matches.metadata. The contract's metadata is going to be stored in this new field. This new field is going to be used by SourcifyDatabaseService as metadata's source of truth instead of RepositoryV2Doubts:
We are going to rebuild production's database from scratch so we are not going to have problems while applyingWe are going to rebuild also staging database from scratchALTER TABLE sourcify_matches ADD COLUMN metadata json NOT NULL. How do we want to handle this case on staging?