Skip to content

build: bump nmilat to v0.2.7, fixing a relay-side PoW mismatch drop - #33

Merged
naliyi merged 5 commits into
mainfrom
bump-nmilat-0.2.7
Sep 4, 2026
Merged

build: bump nmilat to v0.2.7, fixing a relay-side PoW mismatch drop#33
naliyi merged 5 commits into
mainfrom
bump-nmilat-0.2.7

Conversation

@naliyi

@naliyi naliyi commented Sep 4, 2026

Copy link
Copy Markdown
Contributor

Summary

  • Bumps github.com/ohstr/nmilat v0.2.5 → v0.2.7.
  • Fixes a real data-loss bug: any nmilat-backed relay unconditionally rejected an event whose nonce tag overclaimed its NIP-13 difficulty, on ingest, regardless of strictPow/--strict-pow — that ncli setting only ever governed ncli's own read-side verification of events pulled from a from flow, never how the destination relay validated events on write. Fixed upstream in ohstr/nmilat#10, with regression tests that fail on the pre-fix code with the exact reported error and pass on the fix.
  • Also picks up the go1.26.8 minimum-version bump nmilat v0.2.6 already required (go.mod's go directive updated by go mod tidy).

Test plan

  • GOWORK=off go build ./...
  • GOWORK=off go vet ./...
  • GOWORK=off go test -short -race ./... (matches CI, all packages green)
  • GOWORK=off go test ./client/... -run Pow -v — existing strictPow wiring tests unaffected

Upstream fix in ohstr/nmilat#10: relay ingestion unconditionally rejected
any event whose nonce tag overclaimed its NIP-13 difficulty, regardless of
strictPow/--strict-pow (that flag only ever governed ncli's own read-side
check, never the destination relay's). Also picks up the go1.26.8 bump
nmilat v0.2.6 already required.
…now also catches self-contradictory nonce tags

Directly informed by the nmilat v0.2.7 bump: previously any nmilat relay
rejected a self-contradictory nonce tag unconditionally, so this
distinction didn't matter in practice. Now that it's gated behind
Limitation.StrictPow, the relay-side pow.strict doc needed to say what it
actually covers, and apply's strictPow doc needed the caveat that it never
reached the destination relay's own decision anyway -- the actual root
cause of the original bug report.
@naliyi
naliyi marked this pull request as ready for review September 4, 2026 08:20
@naliyi
naliyi merged commit 7113913 into main Sep 4, 2026
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant