[WIP/DRAFT] RFC-49 contract strip graft (ciphertext→catalog) — build-green starting point, test reconciliation INCOMPLETE#1202
Closed
branarakic wants to merge 1 commit into
Conversation
…nt half (WS-A..E) Cherry-pick of 68b80e1 ("feat(rfc49): catalog-sampling rewire + ciphertext strip") from feat/rfc49-catalog-sampling-strip (#1196) onto the Track-C-reconciled agent half. The on-chain random-sampling proof target switches from the private ciphertext commitment (ciphertextChunksRoot) to the public _catalog commitment (catalogRoot); member-side encryption STAYS (ciphertext distributed to members), only the on-chain commitment + core ACK + pricing move to the catalog. REQUIRES A COORDINATED CONTRACT REDEPLOY at an epoch boundary (new addresses + clearOutstandingChallenges migration) — do NOT activate without it. This commit makes the branch LEAD main on the cutover (main still commits ciphertext). Conflict resolution (7 textual + integration fixes the divergence audit missed): - 5 take-strip (ours==main, strip is the catalog cutover): proto/publish-intent.ts, ack-collector.ts, publisher.ts core, storage-ack-handler.ts; deleted the V2 chunked-ciphertext test (feature removed by the strip). - dkg-agent.ts: ciphertext→catalog commitment param swap (chunkedCommitment → catalogCommitment { catalogRoot, catalogLeafCount }). - dkg-publisher.ts: HAND-MERGE — kept ours' curator-ack gate + projection emission (auto-merged regions), grafted the strip's catalog-commitment staging for the curated-publish payload (take-strip on the one staging hunk). - publisher.ts: RESTORED ours+main's skipContextGraphEnsure on PublishOptions (taking strip wholesale had regressed it — strip predates that field). - publisher-runner.ts (cli async-lift, ours-post-strip-base — strip never saw it): renamed the executor's commitment param/forward ciphertext→catalog to match the strip's ACKCollector catalogCommitment param. Verified: ALL 10 packages build green (tsc cross-package type-consistency holds); evm-module compiles 9 contracts. Tests + adversarial review of the dkg-publisher.ts hand-merge pending. Stacked under the agent-half PR (#1201); contract-strip PR is #1196. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Contributor
Author
|
Superseded by the properly-reconciled full cutover branch |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What this is
A cherry-pick of
68b80e18a(the strip commit from #1196, "catalog-sampling rewire + ciphertext strip WS-A..E") onto the Track-C-reconciled agent half. The on-chain random-sampling proof target moves from the private ciphertext commitment (ciphertextChunksRoot) to the public_catalogcommitment (catalogRoot); member-side encryption stays (ciphertext for members), only the on-chain commitment + core ACK + pricing move to the catalog.State (honest)
dkg-publisher.tshand-merge (curator-ack gate preserved) + the strip's storage-ACK gate inversion.packages/core/test/v10-proto.test.ts— stale LU-11 ciphertext test (1 core test fails).packages/publisher/test/storage-ack-handler.test.ts— 157 ours-unique lines a naive take-strip would delete.packages/agent/src/dkg-agent-swm-host.ts— 17 ours-unique curated-detection lines.packages/cli/src/publisher-runner.ts— ours's async-lift path the strip never saw (already renamed here).Why it's harder than a clean graft
The divergence audit projected "7 mechanical conflicts, take-strip, ~1.5 days." That under-counted: main evolved the shared files substantially since the strip's base, so "OURS==main → take strip" regresses main's post-strip additions (e.g.
skipContextGraphEnsure). The correct path is a careful per-file 3-way merge (base = the #1164 common ancestor) + full CI — ideally done with the strip author (#1196).Gating
clearOutstandingChallenges) — this branch makes the line lead main on the cutover.Recommendation
Treat this as scaffolding for the cutover, not a merge candidate. Finish the 3-way reconciliation + full CI with the #1196 author, then un-draft once the redeploy is scheduled.
🤖 Generated with Claude Code