chore(anvil): migrate cheatsmanager to alloy
#6767
Conversation
| @@ -790,7 +794,7 @@ impl TypedTransaction { | |||
| /// Returns true if the transaction was impersonated (using the impersonate Signature) | |||
| #[cfg(feature = "impersonated-tx")] | |||
| pub fn is_impersonated(&self) -> bool { | |||
| self.signature() == IMPERSONATED_SIGNATURE | |||
| to_alloy_signature(self.signature()) == IMPERSONATED_SIGNATURE | |||
There was a problem hiding this comment.
make self.signature() return an alloy signature imo, any reason why we cannot port the typed transaction fields to be alloy?
There was a problem hiding this comment.
thought about doing this, but migrating typedtransaction so early would make us migrate most of anvil in one go with a longer running pr—ideally we want to avoid this.
TypedTransaction, along with other base types are getting migrated on this PR: #6778
| @@ -1,7 +1,8 @@ | |||
| //! Support for "cheat codes" / bypass functions | |||
|
|
|||
| use alloy_primitives::Address; | |||
| use alloy_rpc_types::Signature; | |||
There was a problem hiding this comment.
@DaniPopes are we mixing types here? or should we pull signature into primitives?
There was a problem hiding this comment.
We do have alloy_primitives::Signature! we'll pull it in once needed when we get into actually moving types that can recover their signer (TypedTransaction, e.g) here #6778. For now we can use this type, as right now we just need an equivalence check
Motivation
Part of #6715.
Solution
Migrates the cheats manager and pushes any conversions straight to the signing step, which will then be removed once signers are migrated. While it doesn't remove many conversions, it makes easier to remove them later with the new signers.