fix(SortitionModule): fix staking logic and remove instant staking #2004
Conversation
WalkthroughThis update refactors staking and penalty logic across arbitration contracts by removing the deprecated Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant KlerosCore
participant SortitionModule
User->>KlerosCore: setStakeBySortitionModule(account, courtID, newStake, _)
KlerosCore->>SortitionModule: setStake(account, courtID, newStake, false)
SortitionModule-->>KlerosCore: (StakingResult)
KlerosCore->>SortitionModule: updateState(account, courtID, deposit, withdrawal, newStake)
sequenceDiagram
participant KlerosCore
participant SortitionModule
KlerosCore->>SortitionModule: penalizeStake(account, relativeAmount)
SortitionModule-->>KlerosCore: (pnkBalance, availablePenalty)
KlerosCore->>KlerosCore: update penalty accounting, set juror inactive if needed
sequenceDiagram
participant User
participant SortitionModule
User->>SortitionModule: withdrawLeftoverPNK(account)
SortitionModule-->>User: Transfer leftover PNK, emit event
Possibly related PRs
Suggested labels
Suggested reviewers
Poem
📜 Recent review detailsConfiguration used: CodeRabbit UI 📒 Files selected for processing (8)
✅ Files skipped from review due to trivial changes (1)
🚧 Files skipped from review as they are similar to previous changes (5)
⏰ Context from checks skipped due to timeout of 90000ms (14)
🔇 Additional comments (15)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
CodeRabbit Configuration File (
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🔭 Outside diff range comments (1)
contracts/src/arbitration/university/SortitionModuleUniversity.sol (1)
238-248:⚠️ Potential issueFunction signature updated but missing return values
While the function signature has been updated to match the interface, the function doesn't actually set or return the declared values
pnkBalanceandavailablePenalty. This needs to be fixed to properly implement the interface.Update the function to return the required values:
function penalizeStake( address _account, uint256 _relativeAmount ) external override onlyByCore returns (uint256 pnkBalance, uint256 availablePenalty) { Juror storage juror = jurors[_account]; if (juror.stakedPnk >= _relativeAmount) { juror.stakedPnk -= _relativeAmount; + availablePenalty = _relativeAmount; } else { + availablePenalty = juror.stakedPnk; juror.stakedPnk = 0; // stakedPnk might become lower after manual unstaking, but lockedPnk will always cover the difference. } + pnkBalance = juror.stakedPnk; + return (pnkBalance, availablePenalty); }
🧹 Nitpick comments (12)
contracts/src/arbitration/university/SortitionModuleUniversity.sol (1)
264-265: Empty implementation of withdrawLeftoverPNKThis is an empty implementation of the required interface function. While technically compliant with the interface, this implementation doesn't actually withdraw any leftover PNK. If this is intentional for the University module, consider adding a comment explaining why.
Consider adding a clarifying comment:
-function withdrawLeftoverPNK(address _account) external override {} +function withdrawLeftoverPNK(address _account) external override { + // No implementation needed for University module as it doesn't track leftover PNK +}contracts/src/arbitration/KlerosCoreBase.sol (2)
474-482: Deprecate the unused_alreadyTransferredparameter to save deployment & call gas
setStakeBySortitionModulekeeps the parameter in the ABI but never uses it (/* _alreadyTransferred */).
If no externally-deployed SortitionModule still relies on the 4-argument signature, consider removing the parameter entirely (and adjusting the interface) or adding a dedicated deprecated overload to keep backwards compatibility. Eliminating the dead parameter shaves 3-5 gas per call and simplifies the API.
1083-1099: Follow-up: remove deprecated boolean in call tosetStake
sortitionModule.setStake(..., false)passes a hard-coded value that is now unused.
After cleaning up the interface (see comment above), this fourth argument and the trailing comment can be dropped to avoid confusion:- (uint256 pnkDeposit, uint256 pnkWithdrawal, StakingResult stakingResult) = sortitionModule.setStake( - _account, - _courtID, - _newStake, - false // Unused parameter. - ); + (uint256 pnkDeposit, uint256 pnkWithdrawal, StakingResult stakingResult) = + sortitionModule.setStake(_account, _courtID, _newStake);contracts/test/foundry/KlerosCore.t.sol (5)
1019-1026: Clarify assertion description to match the expected flag value
alreadyTransferredis asserted to befalse, yet the accompanying message says “Should be flagged as transferred”.
This wording is confusing because a value offalseactually means “not transferred”. Renaming the
message to something like “Should be not flagged as transferred” (or the opposite, depending on the intended
semantics) will prevent mis-interpretation when the test fails.
1094-1119: Replace magic numbers with expressions for better readability & resilienceThe balance assertions use large hard-coded literals such as
999999999999998200.
Although they are numerically correct (1 ether - 1800), the intent is obscured and any future change
to the staked amount will require hunting down and editing multiple constants.Consider computing the expected value on-the-fly, e.g.
uint256 expected = 1 ether - 1_800; assertEq(pinakion.balanceOf(staker1), expected, "Wrong token balance of staker1");or even
assertEq( pinakion.balanceOf(staker1), 1 ether - (1500 /* first stake */ + 300 /* diff */), "Wrong token balance of staker1" );This keeps the arithmetic obvious, avoids copy-paste mistakes, and makes the test less brittle.
1203-1210: Impersonating the Core to self-transfer can hide real-world constraints
vm.prank(address(core)); pinakion.transfer(governor, 10000);works in the
unit test, but in productionKlerosCorehas no such method and cannot trigger ERC-20transfer
directly. If the goal is only to zero the Core’s balance, consider minting a fresh token or using
deal(address(pinakion), address(core), 0)(forge cheat-code) instead.This keeps the test closer to real behaviour and avoids accidentally relying on “the contract calls
transfer on itself” paths that will never exist on-chain.
2454-2510: Hard-coded iteration counts & stake figures make new tests brittle
core.execute(disputeID, 0, 6)assumes that exactly six repartition iterations are always required.
The actual number depends on internal constants such asDEFAULT_NB_OF_JURORS, future changes to
penalty logic, or even gas-cap tweaks. The same applies to the literal3000penalty math and the
final balance check for a full1 ether.Suggestion:
uint256 iterations = DEFAULT_NB_OF_JURORS * 2; // worst-case upper bound core.execute(disputeID, 0, iterations);and compute expected balances from recorded pre-/post-state instead of fixed literals.
Doing so will prevent spurious failures when business rules evolve while preserving
the intent of proving “juror becomes insolvent and is automatically unstaked”.
2532-2574: Leverage helper functions to minimise duplicated balance assertions
withdrawLeftoverPNKis tested with a long sequence ofassertEqcalls that
repeat the “balance before / after” pattern already present in several other stake-related tests.
Extracting a small internal helper:function _assertPnkBalances( uint256 coreBal, uint256 jurorBal, string memory msgSuffix ) internal { assertEq(pinakion.balanceOf(address(core)), coreBal, string.concat("Core ", msgSuffix)); assertEq(pinakion.balanceOf(staker1), jurorBal, string.concat("Juror ", msgSuffix)); }and re-using it across the suite will reduce noise, ease future refactors, and make the intent of
each check clearer.contracts/src/arbitration/SortitionModuleBase.sol (4)
69-72: Consider explicitly annotating deprecated storage to reduce cognitive overhead
latestDelayedStakeIndexis kept for storage-layout compatibility, but it now serves no functional purpose.
Adding an/// @deprecatedNatSpec tag (or// DEPRECATED: kept for storage-layout) immediately above the declaration would make the intent clearer to future maintainers and external auditors.
84-100: Mark legacy events with @deprecated to avoid accidental reuseThe three
StakeDelayed*events are retained but unused. Emitting them elsewhere by mistake would silently revive dead semantics.
Consider:-/// @notice DEPRECATED Emitted when a juror's stake is delayed and tokens are not transferred yet. +/// @deprecated Replaced by `StakeDelayed`. Kept for ABI compatibility with historic logs.Doing so on all three legacy events keeps the ABI unchanged while signalling that integrators should subscribe only to
StakeDelayed.
242-253: Guard against zero-iteration calls inexecuteDelayedStakesPassing
_iterations == 0wastes gas and returns silently. A tiny require improves UX and prevents accidental no-op transactions:function executeDelayedStakes(uint256 _iterations) external { require(phase == Phase.staking, "Should be in Staking phase."); require(delayedStakeWriteIndex >= delayedStakeReadIndex, "No delayed stake to execute."); + require(_iterations > 0, "Iterations must be > 0");
411-423: Minor polish forwithdrawLeftoverPNK
The function is callable by anyone for any
_account, which is harmless but unusual.
If the goal is to let third parties “dust-sweep” for jurors, document this explicitly; otherwise considerrequire(msg.sender == _account, "Only juror");.Replace the
if / else { revert(...) }pattern with a singlerequireto save gas and keep style consistent:-function withdrawLeftoverPNK(address _account) external override { +function withdrawLeftoverPNK(address _account) external override { Juror storage juror = jurors[_account]; - if (juror.stakedPnk > 0 && juror.courtIDs.length == 0 && juror.lockedPnk == 0) { - ... - } else { - revert("Not eligible for withdrawal."); - } + require( + juror.stakedPnk > 0 && juror.courtIDs.length == 0 && juror.lockedPnk == 0, + "Not eligible for withdrawal" + ); + ... }
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (7)
contracts/src/arbitration/KlerosCoreBase.sol(4 hunks)contracts/src/arbitration/SortitionModuleBase.sol(8 hunks)contracts/src/arbitration/SortitionModuleNeo.sol(2 hunks)contracts/src/arbitration/dispute-kits/DisputeKitClassicBase.sol(2 hunks)contracts/src/arbitration/interfaces/ISortitionModule.sol(2 hunks)contracts/src/arbitration/university/SortitionModuleUniversity.sol(2 hunks)contracts/test/foundry/KlerosCore.t.sol(10 hunks)
⏰ Context from checks skipped due to timeout of 90000ms (3)
- GitHub Check: Analyze (javascript)
- GitHub Check: SonarCloud
- GitHub Check: contracts-testing
🔇 Additional comments (9)
contracts/src/arbitration/dispute-kits/DisputeKitClassicBase.sol (1)
609-630: Simplified juror eligibility check improves code clarityThe
_postDrawCheckfunction has been simplified to only check if the juror has already been drawn whensingleDrawPerJuroris enabled, removing previous stake sufficiency checks. This aligns with the broader refactoring to simplify staking logic across the arbitration system.contracts/src/arbitration/interfaces/ISortitionModule.sol (2)
30-34: Enhanced penalty accounting with detailed return valuesThe updated return signature for
penalizeStakenow provides more detailed information by returning both the juror's updated PNK balance and the actual penalty applied. This change enables more precise accounting of penalties in the core arbitration contract.
52-53: New function for withdrawing leftover PNK tokensThis new function enables jurors to withdraw any remaining PNK tokens after fully unstaking and clearing their locked tokens, improving the user experience by ensuring no funds are trapped in the contract.
contracts/src/arbitration/SortitionModuleNeo.sol (3)
87-92: Properly marking deprecated parameterThe
_alreadyTransferredparameter has been marked as unused with a comment, clearly indicating its deprecation as part of removing instant staking functionality.
97-104: Simplified conditional logic for stake increasesThe conditional logic has been streamlined to only check maximum stake limits when the stake is being increased, making the code more focused and easier to understand.
112-117: Explicit handling of deprecated parameterThe code now explicitly passes
falsefor the deprecated_alreadyTransferredparameter with a clarifying comment, ensuring proper handling while maintaining compatibility with the base contract.contracts/src/arbitration/KlerosCoreBase.sol (1)
789-803: Double-unlock edge case: verify that unlocked but un-penalised PNK cannot be withdrawnThe flow is:
unlockStake(account, penalty);penalizeStake(account, penalty)→ returnsavailablePenalty(may be < penalty).If the juror’s locked stake is smaller than
penalty, the surplus tokens get unlocked in step 1 but are not slashed in step 2, letting the juror immediately withdraw them-–potentially rewarding under-collateralised behaviour.Please confirm that
sortitionModule.unlockStakeprevents unlocking more than currently locked or thatpenalizeStakeinternally adjusts the unlocked amount. If not, swap the order (penalise first, then unlock the actualavailablePenalty) or unlock onlyavailablePenalty.contracts/src/arbitration/SortitionModuleBase.sol (2)
327-334: Double-check tree consistency when withdrawals are limited by locked PNKIf
lockedPnkprevents a full withdrawal,pnkWithdrawalis trimmed but_newStakeremains unchanged.
This deliberately produces “left-over” PNK that can later be withdrawn viawithdrawLeftoverPNK, so it may be correct by design.
However, it temporarily breaks the invariantjuror.stakedPnk == Σ(tree stakes)until the leftover is withdrawn.Please verify that all downstream logic (drawing, penalties, etc.) tolerates this mismatch; otherwise consider updating
_newStakeafter trimming:- // _newStake is still the user-requested target (might be < effective stake) + // Align stake recorded in the tree with the actual post-withdrawal amount. + _newStake = currentStake - pnkWithdrawal;
372-389: Potential invariant break:lockedPnkcan exceedstakedPnkafter penalties
penalizeStakereducesstakedPnkbut never toucheslockedPnk.
IflockedPnk > stakedPnkafterwards, subsequent calls tounlockStakecould underflow or leavestakedPnknegative in intent (though uint256 prevents it).Please confirm that
lockedPnk ≤ stakedPnkis not relied upon elsewhere (e.g., when drawing, displaying balances, or executing further penalties).
An inexpensive safeguard could be:assert(juror.stakedPnk >= juror.lockedPnk);or clamping the penalty to
stakedPnk - lockedPnk.
| /// @dev Transfers PNK to the juror by SortitionModule. | ||
| /// @param _account The account of the juror whose PNK to transfer. | ||
| /// @param _amount The amount to transfer. | ||
| function transferBySortitionModule(address _account, uint256 _amount) external { | ||
| if (msg.sender != address(sortitionModule)) revert SortitionModuleOnly(); | ||
| // Note eligibility is checked in SortitionModule. | ||
| pinakion.safeTransfer(_account, _amount); | ||
| } |
There was a problem hiding this comment.
🛠️ Refactor suggestion
Emit an event for PNK moved via transferBySortitionModule
transferBySortitionModule performs an external token transfer but leaves no on-chain trace inside Core.
For transparency, accounting and easier off-chain indexing, emit a lightweight event such as:
+ event SortitionPNKTransfer(address indexed account, uint256 amount);
...
function transferBySortitionModule(address _account, uint256 _amount) external {
if (msg.sender != address(sortitionModule)) revert SortitionModuleOnly();
pinakion.safeTransfer(_account, _amount);
+ emit SortitionPNKTransfer(_account, _amount);
}This incurs negligible gas yet gives jurors and auditors full visibility.
🤖 Prompt for AI Agents
In contracts/src/arbitration/KlerosCoreBase.sol around lines 483 to 490, the
function transferBySortitionModule transfers PNK tokens without emitting any
event, which reduces transparency and traceability. Add an event declaration for
PNK transfers by the SortitionModule and emit this event inside the function
after the safeTransfer call, including the recipient address and the amount
transferred to provide on-chain visibility and facilitate off-chain indexing.
✅ Deploy Preview for kleros-v2-testnet ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
❌ Deploy Preview for kleros-v2-university failed. Why did it fail? →
|
✅ Deploy Preview for kleros-v2-testnet-devtools ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
✅ Deploy Preview for kleros-v2-neo ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
jaybuidl
added
Type: Bug 🐛
Compatibility: ABI change 🗯
There was a problem hiding this comment.
Actionable comments posted: 0
🔭 Outside diff range comments (1)
subgraph/core/abi-migrations/SortitionModule.json (1)
823-843: 💡 Verification agent🧩 Analysis chain
Invalid unnamed boolean input in setStake
The input list forsetStakestill contains an unnamedboolparameter (formerly_alreadyTransferred), which is invalid in the ABI and will break code generation. You should either remove this entry entirely or assign it a proper name to match the contract signature.Apply this diff to remove it:
--- before { "internalType": "bool", - "name": "", + "name": "", // <–– remove this block if the parameter is deprecated "type": "bool" } --- after // block removed
🏁 Script executed:
#!/usr/bin/env bash # Extract the full ABI entry for the `setStake` function to verify its inputs jq '.[] | select(.name == "setStake")' subgraph/core/abi-migrations/SortitionModule.jsonLength of output: 198
🏁 Script executed:
#!/usr/bin/env bash # Attempt to extract the `setStake` entry from the ABI jq '.abi[] | select(.name=="setStake")' subgraph/core/abi-migrations/SortitionModule.json \ || rg -n '"name": "setStake"' -A 10 -B 10 subgraph/core/abi-migrations/SortitionModule.jsonLength of output: 984
🏁 Script executed:
#!/usr/bin/env bash # Locate the Solidity definition of setStake to confirm the parameter name rg -n "function setStake" -g "*.sol" -C 3Length of output: 6473
Remove Invalid Unnamed Boolean from
setStakeABI
The ABI forsetStakecurrently includes a fourth, unnamedboolparameter that isn’t part of the contract’s external signature (address,uint96,uint256only). This extra input breaks codegen and must be removed.• File:
subgraph/core/abi-migrations/SortitionModule.json(inputs around lines 823–843)
• Function:setStake(address _account, uint96 _courtID, uint256 _newStake)Apply this diff to drop the stray boolean block:
--- a/subgraph/core/abi-migrations/SortitionModule.json +++ b/subgraph/core/abi-migrations/SortitionModule.json @@ -823,10 +823,6 @@ { "internalType": "uint256", "name": "_newStake", "type": "uint256" }, - { - "internalType": "bool", - "name": "", - "type": "bool" - }, { "internalType": "uint256", "name": "pnkDeposit", "type": "uint256" },
🧹 Nitpick comments (2)
subgraph/core/abi-migrations/SortitionModule.json (2)
67-81: Added LeftoverPNKWithdrawn event
Introduces theLeftoverPNKWithdrawn(address _account, uint256 _amount)event for leftover PNK withdrawals. Make sure your subgraph’s GraphQL schema and mapping handlers are updated to index this new event.
954-961: New withdrawLeftoverPNK function
AddswithdrawLeftoverPNK(address _account)for direct PNK withdrawals. Update any subgraph custom handlers or scripts that invoke this function if you intend to index these calls.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (6)
subgraph/core-neo/abi-migrations/SortitionModuleNeo.json(6 hunks)subgraph/core-neo/subgraph.yaml(1 hunks)subgraph/core/abi-migrations/SortitionModule.json(7 hunks)subgraph/core/src/SortitionModule.ts(1 hunks)subgraph/core/subgraph.yaml(1 hunks)subgraph/temp-older-events-addition.txt(1 hunks)
✅ Files skipped from review due to trivial changes (1)
- subgraph/temp-older-events-addition.txt
⏰ Context from checks skipped due to timeout of 90000ms (7)
- GitHub Check: Redirect rules - kleros-v2-testnet-devtools
- GitHub Check: Header rules - kleros-v2-testnet-devtools
- GitHub Check: Pages changed - kleros-v2-testnet-devtools
- GitHub Check: Redirect rules - kleros-v2-testnet
- GitHub Check: Header rules - kleros-v2-testnet
- GitHub Check: Pages changed - kleros-v2-testnet
- GitHub Check: contracts-testing
🔇 Additional comments (14)
subgraph/core/subgraph.yaml (1)
162-163: LGTM! Event consolidation simplifies the handling logic.The consolidation of multiple
StakeDelayedevent variants into a single unified event handler aligns well with the contract changes that removed the deprecated_alreadyTransferredparameter. This simplification reduces complexity and maintains consistency across the system.subgraph/core-neo/subgraph.yaml (1)
161-162: LGTM! Consistent event handling across networks.The changes mirror those in the core network subgraph, ensuring consistent event handling across both arbitrum-sepolia and arbitrum-one networks. The unified
StakeDelayedevent signature maintains compatibility while simplifying the codebase.subgraph/core/src/SortitionModule.ts (2)
1-1: LGTM! Import statement correctly reflects the consolidated events.The import statement now includes only the necessary events after the consolidation, removing the deprecated variants. This keeps the imports clean and aligned with the contract changes.
7-9: LGTM! Unified handler simplifies event processing.The single
handleStakeDelayedfunction correctly extracts all necessary parameters (_address,_courtID,_amount) from the consolidated event. This approach eliminates code duplication while maintaining full functionality.subgraph/core-neo/abi-migrations/SortitionModuleNeo.json (6)
69-80: LGTM! Event rename clarifies its purpose.The rename from
StakeDelayedAlreadyTransferredDepositedtoLeftoverPNKWithdrawnmakes the event name more descriptive and aligns with its actual functionality of withdrawing leftover PNK tokens.
87-93: LGTM! Simplified NewPhase event reduces complexity.The
NewPhaseevent now only includes the essential_phaseparameter as auint8enum, removing unnecessary court ID and amount parameters that were previously included. This simplification makes the event more focused and easier to handle.
117-118: LGTM! Consolidated StakeDelayed event improves consistency.The unified
StakeDelayedevent with indexedaddressanduint96court ID parameters, along with theuint256amount, provides a clean interface that eliminates the need for multiple event variants.
742-753: LGTM! Enhanced penalizeStake function provides better accountability.The updated
penalizeStakefunction now returns bothpnkBalanceandavailablePenaltyvalues, which improves transparency by providing information about the juror's remaining balance and the actual penalty applied. This is valuable for accurate penalty accounting.
885-905: LGTM! setStake function improvements enhance functionality.The changes to
setStakeinclude:
- Unnamed boolean parameter (formerly
_alreadyTransferred) indicating deprecation- Additional
oldStakereturn value for better state trackingThese changes improve the function's utility while maintaining backward compatibility through the unnamed parameter.
1024-1037: LGTM! New withdrawLeftoverPNK function adds important functionality.The addition of
withdrawLeftoverPNKfunction provides jurors with a mechanism to withdraw any remaining PNK tokens after full unstaking. This is an important feature for complete token recovery and aligns with the overall staking logic improvements.subgraph/core/abi-migrations/SortitionModule.json (4)
4-7: Explicit no-args constructor declared
The ABI now includes a parameterless constructor (type: "constructor") and has removed any fallback/receive entries to align with the updated contract deployment.
87-93: Consolidated NewPhase event
Replaces legacy phase‐related events with a singleNewPhase(uint8 _phase)event. Verify that your subgraph mappings now listen toNewPhaseinstead of the old events.
114-118: Updated StakeDelayed event
All delayed‐stake events have been merged intoStakeDelayed(address _address, uint96 _courtID, uint256 _amount). Ensure your subgraph only handles this consolidated event going forward.
679-691: penalizeStake now returns actual balances
ThepenalizeStakefunction signature was changed to return(uint256 pnkBalance, uint256 availablePenalty). If your subgraph makes contract calls to this function, update the ABICall definitions accordingly.
unknownunknown1
force-pushed
the
fix/staking-fix
branch
from
d128f10 to
f7591d2
Compare
|
Code Climate has analyzed commit f7591d2 and detected 0 issues on this pull request. View more on Code Climate. |
|
PR-Codex overview
This PR focuses on enhancing the
SortitionModulefunctionality by introducing new methods for managing stakes and penalties, improving the handling of delayed stakes, and adding a mechanism for withdrawing leftover tokens. It also includes modifications to existing functions for better clarity and efficiency.Detailed summary
Delayedstate toStakingResultenum.penalizeStakefunction to returnpnkBalanceandavailablePenalty.withdrawLeftoverPNKfunction for jurors to withdraw unused tokens.updateStatefunction to manage stake updates._alreadyTransferredparameter from several functions.Summary by CodeRabbit
New Features
Improvements
Bug Fixes
Tests