deployment: base and mainnet #50
Conversation
Signed-off-by: Elliot <elliotfriedman3@gmail.com>
Signed-off-by: Elliot <elliotfriedman3@gmail.com>
Signed-off-by: Elliot <elliotfriedman3@gmail.com>
Signed-off-by: Elliot <elliotfriedman3@gmail.com>
| keccak256(type(TimelockFactory).runtimeCode), | ||
| keccak256(factory.code.sliceBytes(0, 24116)), | ||
| keccak256( | ||
| type(TimelockFactory).runtimeCode.sliceBytes(0, 24116) |
There was a problem hiding this comment.
there is differences in last 128 bytes. Works with sliceBytes(0, 24187).
I don't understand why 128 bytes are different when CBOR bytes tell that metadata is 51 bytes long.
There was a problem hiding this comment.
last 138 bytes cab be removed.
a2646970667358221220faf3b7cb28c8bf5fe4d307231fad9edac131eccb00c5a5e363d013efe4a59a9764736f6c6343000819003308f913ffb0ba4369820330664f249be5a609b8d6503893d07eda734fd42bd4c6a264697066735822122090333ae03f10ea21d15a94de29f40350f00e215ed90ab2b63844e3dae2d05b6364736f6c63430008190033
There was a problem hiding this comment.
yeah i'm not sure what accounts for this change either, only that it is necessary to make things work.
| assertEq( | ||
| keccak256(guard.code), | ||
| keccak256(type(Guard).runtimeCode), | ||
| keccak256(guard.code.sliceBytes(0, 950)), |
There was a problem hiding this comment.
there is difference in last 53 bytes. 2 CBOR bytes and 51 metadata bytes. so removing last 53 bytes would work i.e. sliceBytes(0, 953)
| assertEq( | ||
| keccak256(recoverySpellFactory.code), | ||
| keccak256(type(RecoverySpellFactory).runtimeCode), | ||
| keccak256(recoverySpellFactory.code.sliceBytes(0, 9470)), |
There was a problem hiding this comment.
Last 106. bytes are metadata. two metadatas one for RecoverySpell and other for RecoverySpellFactory. So last 106 bytes should be removed.
Ref: ethereum/solidity#14827
There was a problem hiding this comment.
This is the data to be removed.
a26469706673582212207bc965e90185648126b9111f159955c4b4afbc31cf59b1f93a4ec86501d4215e64736f6c63430008190033a2646970667358221220903c7bdd4b158ba0037ba1c0fc25619bd86084c68e73a918d80f5b7310271df864736f6c63430008190033
| @@ -77,23 +80,29 @@ contract SystemDeploy is MultisigProposal { | |||
| if (addresses.isAddressSet("TIMELOCK_FACTORY")) { | |||
| address factory = addresses.getAddress("TIMELOCK_FACTORY"); | |||
| assertEq( | |||
| keccak256(factory.code), | |||
| keccak256(type(TimelockFactory).runtimeCode), | |||
| keccak256(factory.code.sliceBytes(0, 24116)), | |||
There was a problem hiding this comment.
I thought .code only returns the runtime bytecode. What is the reason for the slice here? Could you add a comment explaining, please? And why at 24116? How did you calculate that?
There was a problem hiding this comment.
It was calculated based on the data that afterwards we needed to remove.
There was a problem hiding this comment.
The only thing we can get from a deployed contract is the bytecode as it exists onchain. So yes, that is what we are getting here.
|
Going to redeploy things, will see if the metadata issue persists on redeployment, and if so will slice the last bytes off. |
…ainnet-test-deployment
No description provided.