task_240926_16 — Dispute Seed (#151), Treasury Forwarding (#152), Indexer Events (#153) (Robbert)

[umeradl]

cc @robertocarlous
Assigned. Full spec: Developer/tasks/task_240926_16.md

The task file has the full specification: exact file/line references, per-part background, and the deliverables checklists. This post is the summary, the assignment notice, and the decisions that need flagging while you build.

Timeline

7 days: Sep 24 – Oct 1, 2026. This window also covers the follow-on task_17, which will be written and linked here shortly. Budget Parts A–C to leave room for it.

Summary

Three independent pieces of task-contract work, submitted as three separate PRs into develop in order A → B → C, each rebased after the previous one merges. Each part implements the scope agreed in its issue's review comment, not the original issue text:

• Part A: #151 / BL-11, ungrindable dispute-subgroup seed.
  • Replace _assignFreshSubgroup's blockhash(block.number - 1) with a future-block seed anchored at openDispute, locked by a permissionless lockDisputeSeed, with a re-anchor path if 256 blocks pass.
  • _shuffleAddressArray takes an explicit seed so security: aggregator batch shuffle uses grindable blockhash + no commit-reveal on T1/T2 submissions (H-2/M-1 aggregation-side) #156 can reuse it.
  • This supersedes the challenger commit-reveal in the issue (review).
• Part B: #152, platform-resolved treasury forwarding.
  • Remove the model-owner-settable setTreasuryAddress from both task contracts, and resolve the treasury via DinValidatorStake.slashTreasury().
  • Route all six forfeiture/penalty/fee paths through one helper that burns if no treasury is set. That fixes four paths where tokens are currently stranded (review).
• Part C: #153, task-level lifecycle events for the indexer.
  • The six events from PR feat(indexer): implement DIN Protocol subgraph — platform + task-level contracts #29's audit §4 (including T1BatchFinalized/T2Finalized, which the issue missed), plus a _setGIstate helper that emits the new GI on GIstarted.
  • The subgraph half lands as commits on PR feat(indexer): implement DIN Protocol subgraph — platform + task-level contracts #29 after PR feat(indexer): wire P3-staking events — ValidatorJailed/Reactivated, governable MIN_STAKE, JailEvent entity #72 is folded into it (review, Discussion #162).

Decisions to flag, not guess

1. Part A, disputeSeedDelay default: placeholder in the 5–10 block range. Pick one and justify it in the PR description.
2. Part A, design: if you find a reason to prefer commit-reveal or VRF over the future-block seed once you're in the code, raise it on BL-11: Replace blockhash randomness in _assignFreshSubgroup with commit-reveal #151 before building it.
3. Part B, settleRewards fee share: it stays forwarded in full through a forward-only helper, since it's a protocol fee, not a forfeiture. Only forfeitures and penalties get the 50/50 burn/treasury split. Raise it on Wire treasury forwarding from DINTaskCoordinator and DINTaskAuditor to DinTreasury #152 first if you disagree.
4. Part B, helper location: per-contract or DINShared.sol, your call. Say which in the PR.
5. Part C, contract size: put forge build --sizes before/after numbers in the PR. If DINTaskCoordinator crosses the 24 KB EIP-170 limit, stop and raise it here rather than splitting the contract.
6. Part C, ETHTreasuryWithdrawn stub on the subgraph: drop it, or repoint it at DinTreasury's actual events (EthReceived/EthWithdrawn/ERC20Withdrawn). Your call; state it in feat(indexer): implement DIN Protocol subgraph — platform + task-level contracts #29's description.

Dependency

Part C0 (fold #72 into #29, then refresh against develop) is a prerequisite for Part C's subgraph half only, not for C1–C4. The upstream feat/din-indexer base has already been fast-forwarded to develop (#162 note). Steps 1 and 3 there are yours.

Reference material

• Developer/tasks/task_240926_16.md: full spec
• #151, #152, #153: tracking issues (each has a review comment with the agreed scope)
• Discussion #162: indexer-branch consolidation
• Developer/BACK_LOG.md BL-11
• Developer/design/MECHANISM_DESIGN.md §4 (slashing/disputes), §8 (fees/treasury)

Post progress updates, questions, and PR links here as the task moves, same convention as prior task-tracking discussions (e.g. #137, #138). If anything in the task spec doesn't feel right, is ambiguous, or needs clarification, comment here and cc @umeradl.

[umeradl]

@robertocarlous: task_17 is now written and assigned. It's tracked separately in Discussion #164, and the full spec is in Developer/tasks/task_240926_17.md.

It covers the P3-6.2 adversarial suite (#154: threat-model doc first, then P3Adversarial.t.sol) and testnet-parameter readiness (#155: env-parameterised DeployPlatform.s.sol, the 0-bps S1/S2 slash-fraction fix, and the DinEmission NatSpec fix; the values come from the Fri Sep 25 meeting).

Both tasks share the Sep 24 – Oct 1 window. Keep task_16 updates here and task_17 updates in #164. If the two don't fit together, raise it in either thread.

[robertocarlous]

task_16 complete. Three PRs open against InfiniteZeroFoundation/DevNet develop:

• #171 — feat/issue-151-dispute-seed: BL-11 future-block dispute seed. openDispute anchors seedBlock = block.number + disputeSeedDelay (default 7); permissionless lockDisputeSeed; resolveDispute(upheld=true) gates on locked seed; 256-block re-anchor path. 26/26 DisputeResolutionTest pass.

• #172 — feat/issue-152-treasury-forwarding: Platform-resolved treasury. Removed per-task setTreasuryAddress/treasuryAddress from both task contracts; treasury now resolved via IDinValidatorStake.slashTreasury() at payout time. _burnAndForward (50/50) for forfeitures, _forwardToTreasury (100%) for protocol fees. All 6 stranded paths wired. 12/12 TreasuryForwardingTest pass.

• #173 — feat/issue-153-lifecycle-events: Six GI lifecycle events (GIStateChanged, LocalModelSubmitted, T1AggregationSubmitted, T2AggregationSubmitted, T1BatchFinalized, T2Finalized). _setGIstate helper replaces 23 bare state assignments; fixes GI-index ordering on GIstarted (GI increments before emit). ABIs refreshed. 9/9 LifecycleEventsTest pass.

All three target develop on the upstream repo.

[robertocarlous]

cc: @umeradl

[umeradl]

Heads-up: dincli dindao is now dincli dinrep — PR No. 175 landed on develop (merge 5864f7d, follow-up 8f9a90b). The DIN-Representative command group is renamed to match the role: dincli/cli/dindao.py → dincli/cli/dinrep.py, Documentation/public/roles/dindao.md → dinrep.md. Breaking, no alias — dincli dindao … now returns "No such command". Subcommands are unchanged (dincli dinrep deploy …, dincli dinrep add-slasher …, dincli dinrep registry …).

For this task: task_240926_16.md's indexer merge step now notes the rename. PR No. 29 and No. 72 both modify dincli/cli/dindao.py, so merging develop into them will hit a rename/modify conflict: re-apply those edits to dincli/cli/dinrep.py (the file content is otherwise unchanged apart from two help strings), and update the doc references to dindao.py in the indexer docs to dinrep.py.

[umeradl]

@robertocarlous: status check on task_16 against develop (8f9a90b). Parts A, B, and the contract half of C have landed. The task stays open for the subgraph half of Part C.

Landed

Part	PR	Merge	Issue
A: future-block dispute seed (BL-11)	No. 171, merged 2026-09-26	584bb83	No. 151 closed; BL-11 marked resolved in BACK_LOG.md (sequencer-trust residual tracked in No. 178)
B: platform-resolved treasury forwarding	No. 172, merged 2026-09-26	b6304d7	No. 152 closed
C1–C4: lifecycle events + _setGIstate	No. 173, merged 2026-09-27	e9dc3f0	No. 153 closed

Still open: Part C subgraph half (task file, Deliverables (Part C))

• C0: fold PR No. 72 into PR No. 29 (fast-forward), then close No. 72. Refresh No. 29 against develop, regenerate subgraph/abis/, and get graph build passing. No. 72 is still open, and No. 29's head is still 866521a (2026-07-13).
• C5, as commits on No. 29:
  • ABIs regenerated and the six handlers enabled
  • enum mapping regenerated from the current GIstates
  • RewardDeposited handler added
  • ETHTreasuryWithdrawn stub resolved (state the decision in No. 29's description)
  • audit §4.1/§5 and daemon-event-schema.md corrected
• Comment on No. 29 when C0 is done and again when C5 is pushed. Update No. 29's description to cover No. 72's folded-in scope and to reference No. 153.

Heads-up: when you merge develop into No. 29/No. 72, expect a rename/modify conflict on dincli/cli/dindao.py → dinrep.py (see the dinrep note above).

This discussion closes once C0 and C5 are done.

[umeradl]

Status check (2026-09-30): not closing yet. Nothing has changed since the Sep 28 check: the subgraph half of Part C is still open.

Checked against origin/develop at 6ccc28c.

Part	Status	Evidence
A: future-block dispute seed (No. 151, BL-11)	✅ Landed	PR No. 171, merged 2026-09-26 (584bb83, follow-ups 9276d51). No. 151 closed.
B: platform-resolved treasury forwarding (No. 152)	✅ Landed	PR No. 172, merged 2026-09-26 (b6304d7, follow-ups cfe2af4). No. 152 closed.
C1–C4: lifecycle events + _setGIstate (No. 153)	✅ Landed	PR No. 173, merged 2026-09-27 (e9dc3f0, follow-ups 04aa58d). No. 153 closed.
C0: fold PR No. 72 into PR No. 29	⏳ Not started	PR No. 72 is still open. PR No. 29's head is still 866521a (2026-07-13), and feat/din-indexer / feat/din-indexer-65 on your fork haven't moved.
C5: subgraph half, as commits on PR No. 29	⏳ Not started	No new commits on PR No. 29.

My side of C0 is done: the upstream feat/din-indexer base branch is fast-forwarded to current develop (6ccc28c, Sep 30), so it's no longer a stale July base. When you merge develop into your branch, PR No. 29's diff will show only the subgraph work. If you end up merging a newer develop than 6ccc28c, ping me here and I'll fast-forward the base again, otherwise No. 29's diff picks up those develop commits.

@robertocarlous, two things have changed for C5 since the task was written:

• GIstates has 26 members now, not 24. PR No. 197 (task_240926_18 Part C) inserted T1AggregationRevealStarted (18) and T2AggregationRevealStarted (21), so every ordinal from 18 up has shifted again. The full table is in this comment on PR No. 29. Regenerate the enum mapping from current foundry/src/DINShared.sol, not from the task file's numbers.
• The submit functions are gone. submitT1Aggregation/submitT2Aggregation are now commitT1Aggregation/revealT1Aggregation and the T2 equivalents. T1AggregationSubmitted/T2AggregationSubmitted keep their signatures but are emitted on reveal. Regenerate subgraph/abis/ from develop after merging it in.

The dincli/cli/dindao.py → dinrep.py rename conflict from the dinrep note above still applies.

Remaining checklist (unchanged):

• C0: fast-forward your feat/din-indexer to feat/din-indexer-65, close PR No. 72 with a "folded into No. 29" note, merge develop in, regenerate subgraph/abis/, get graph codegen && graph build passing.
• C5, as commits on PR No. 29: six handlers enabled, enum mapping regenerated, RewardDeposited handler added, ETHTreasuryWithdrawn stub resolved, audit §4.1/§5 and daemon-event-schema.md corrected.
• Comment on PR No. 29 when C0 is done and again when C5 is pushed, and update its description (No. 72's folded-in scope, reference to No. 153, the ETHTreasuryWithdrawn decision).

The window ends Oct 1. If C0/C5 won't fit alongside task_17 Part 2, say so here and give a date.

This discussion closes once C0 and C5 are done.