-
Notifications
You must be signed in to change notification settings - Fork 34
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
4844 #157
base: main
Are you sure you want to change the base?
4844 #157
Conversation
src/test/identity-manager/WorldIDIdentityManagerDataQuery.t.sol
Outdated
Show resolved
Hide resolved
src/test/identity-manager/WorldIDIdentityManagerIdentityRegistration.t.sol
Outdated
Show resolved
Hide resolved
src/test/identity-manager/WorldIDIdentityManagerIdentityRegistration.t.sol
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Some style + this is cryptographically broken
440e522
to
c309622
Compare
6ca49e5
to
a5ab7aa
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One minor wording issue
src/WorldIDIdentityManagerImplV3.sol
Outdated
/// @notice The root obtained after inserting all of `identityCommitments` into the tree described | ||
/// by `preRoot`. Must be an element of the field `Kr`. (already in reduced form) | ||
uint256 postRoot; | ||
/// @notice Root of the merkle tree after adding identities to it. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is misleading. Better wording: Hash of all inserted identities, constructed by taking a root of the Merkle Tree of minimal depth containing all identities
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed
Every contract seems to use 4 anyway. Signed-off-by: Wojciech Zmuda <[email protected]>
This is groundwork for EIP-4844 support. Cancun introduces the blobhash opcode and SOLC 0.8.24 is the first revision to support this opcode in Solidity.
Signed-off-by: Wojciech Zmuda <[email protected]>
It does not seem to be used anywhere. Signed-off-by: Wojciech Zmuda <[email protected]>
This is ground work for EIP-4844. Vm.blobhashes allow for mocking value of the next call to the blobhash opcode. This will be useful in EIP-4844 tests to mock this opcode. Update the usage of the one symbol that changed its name. Signed-off-by: Wojciech Zmuda <[email protected]>
Implement WorldID Identity Manager in its 3rd implementation. The new implementation contains one main method - registerIdentities. It's the EIP-4844 equivalent of the existing registerIdentities from the V1 implementation. The new registerIdentities calls verifyProof that is also different from its classic counterpart. The new verifyProof comes from the EIP-4844-ready InsertionTreeVerifier164844 contract, that was generated from the semaphore-mtb's InsertionCircuit modified to accommodae EIP-4844 features. Since the new verifyProof function has a different prototype than the previous one, the ITreeVerifier interface has been updated with the new prototype. This change results in lots of boilerplate changes as all contracts implementing this interface must contain a dummy implementation of the new function, even if it does not make practical sense. Add tests for the new feature. Tests are modelled by the existing registration tests. The new contract is basically a copy of the old one, with the main difference of calling the new registerIdentities implementation. Signed-off-by: Wojciech Zmuda <[email protected]>
Making these variables global shrunks stack significantly, allowing to build contracts without the --via-ir flag. Signed-off-by: Wojciech Zmuda <[email protected]>
The --gas-report flag of the forge test command seems to conflict with blobhash mocks in tests, making the new EIP-4844 tests fail, despite they pass without this flag. Turn off benchmarks temporarily, to unblock EIP-4844 development. This should be debugged and fixed later. Signed-off-by: Wojciech Zmuda <[email protected]>
@Dzejkop @0xKitsune, after consulting @kustosz we decided to temporarily block the benchmark step of the CI. Forge tests go bonkers if |
No description provided.