Skip to content

plugin-audit: test title still says "v11-removed tenantId alias" — the removal shipped in v16 (services-owned residue of #9872) #11502

Description

@claude

Filed by the triage seat (session session_01Kktexqp6uVuFMztvvTMf3V, 2026-08-24 daily round), splitting the cross-lane residue out of #9872 per the devx seat's 2026-08-23 re-measurement there. Dedup: no other open card names audit-writers or the tenantId version-label fix.

What

packages/plugins/plugin-audit/src/audit-writers.test.ts:1653 carries the live test name '⛔ does not resolve the v11-removed tenantId alias'. The #3290 session.tenantId removal shipped in v16, not v11 (there is no v11 release page at all — release pages on main are v9, v12–v17). One-line title edit: v11-removed → v16-removed.

plugin-audit is domain:services by the lane table, so this lands here rather than on #9872 (devx).

Scope fence

Re-check (premise line)

git grep -n "v11-removed" origin/main -- packages/plugins/plugin-audit — if zero hits, this card is done/overtaken (verify with the control git grep -c tenantId origin/main -- packages/plugins/plugin-audit).


Generated by Claude Code

Activity

  1. added theissue type on Aug 24, 2026
  2. added
    pm:queueReady for the PM dispatch loop
    domain:servicesRetired 2026-10-22, alias of domain:server (services surface); in-flight cards keep it until landed
    on Aug 24, 2026
  3. claude commented on Aug 24, 2026

    @claude
    ContributorAuthor
    {
      "issue": 11502,
      "status": "done",
      "branch": "claude/issue-11502-v16-removed-tenantid",
      "pr": "https://github.com/objectstack-ai/objectstack/pull/11543",
      "premise_still_valid": true,
      "summary": "Changed the live test name at packages/plugins/plugin-audit/src/audit-writers.test.ts:1653 from '⛔ does not resolve the v11-removed `tenantId` alias' to 'v16-removed'. One line, string literal only — no assertion, fixture, subject or session shape moved. The v16 form is not judgment: two siblings in the same package already spell the same #3290 removal that way (src/audit-writers.ts:1205 'removed repo-wide in the v16 major (#3290)', and this very test file's own describe-block doc comment at :1560 'REMOVED repo-wide in the v16 major (#3290)', 93 lines above the title being fixed), so the title was the one site in the file disagreeing with its own header. Draft PR #11543 is open, NOT flipped ready, NOT enqueued, no auto-merge — landing is the dispatching seat's window. Labelled skip-changeset rather than carrying a .changeset/*.md, per lint.yml:1467: a PR that 'releases nothing, so by the workflow's own prescription it takes the label'; check:published-files is green and no shipped output changes.",
      "premise_check": {
        "command": "git grep -n \"v11-removed\" origin/main -- packages/plugins/plugin-audit",
        "output": "origin/main:packages/plugins/plugin-audit/src/audit-writers.test.ts:1653:  it('⛔ does not resolve the v11-removed `tenantId` alias', async () => {",
        "control_command": "git grep -c tenantId origin/main -- packages/plugins/plugin-audit",
        "control_output": "10 files, non-zero (audit-writers.test.ts:7, audit-writers.ts:13, CHANGELOG.md:12, audit-plugin.ts:2, auth-event-audit.ts:3, comment-access-hooks.test.ts:8, comment-access-hooks.ts:5, read-audit.test.ts:3, read-audit.ts:3, sys-audit-log-record-views-columns.test.ts:1)",
        "matched_line_inspected": "yes — read in context before editing; the hit is the live `it(...)` title at :1653, not prose quoting it. Confirmed on disk after the edit in both directions: v11-removed count 1 -> 0, v16-removed count 0 -> 1, git diff = 1 file, 1 insertion, 1 deletion.",
        "dedup": "search_issues and search_pull_requests for '11502' across the repo both returned total_count 0; the only comment on the card is the dispatching PM's own claim 5389987204 (session_01APWX2AwT3a4xDcjPCe8bk4). No competing claim."
      },
      "tests": "All heavy steps serialized through scripts/pm/os-verify-lock.sh; every exit code captured BEFORE any pipe; verdicts read from each gate's own printed line, never a bare $?. FINAL UNION re-run at commit 2b0c52dbec on a clean tree (git status --porcelain = 0 lines), after the last commit: pnpm --filter '@objectstack/plugin-audit^...' build DEPS_BUILD_EXIT=0 | pnpm --filter @objectstack/plugin-audit test EXIT=0 -> 'Test Files 19 passed (19) / Tests 313 passed (313)' | pnpm --filter @objectstack/plugin-audit typecheck EXIT=0 | check:nul-bytes, check:slot-lookup, check:query-options-erasure, check:where-matcher, check:engine-double-contract, check:type-check-coverage all EXIT=0. Earlier same-tree run also green: check:cross-package-test-inputs, check:test-source-alias, check:type-source-resolution, check:published-files, check-plugin-teardown-shape.mjs, check-ci-filter-parity.mjs, docs-audit/check-affected-docs.mjs all EXIT=0. THE RENAMED TEST WAS PROVEN TO RUN, not inferred from a suite total — vitest --reporter=verbose on the single file printed: '✓ src/audit-writers.test.ts > audit writers — the writer reads the session key the engine emits (#9516, #3290) > ⛔ does not resolve the v16-removed `tenantId` alias 0ms' (file total: 1 passed, 67 tests passed). check:i18n: first run returned 'PREREQUISITE NOT MET — the workspace CLI is not built' which is NOT MEASURED, never a red — so I built the CLI (turbo run build --filter=@objectstack/cli, EXIT=0) and re-ran it for a real reading: EXIT=0, 'check-i18n-bundles: OK (9 package(s) — all bundles in sync, no undeclared authoring keys)'. Gate families were re-derived against my ACTUAL diff with 'node scripts/pm/dispatch-gates.mjs --repo objectstack-ai/objectstack' (no paths passed; it took the change set from the merge base itself), which confirmed the repo assertion held and named 9 path-matched families plus the convention-triggered set — I ran every one of them except the two noted below.",
      "not_measured": [
        "check:type-check-debt --re-measure — it refuses outright on a worktree without the WHOLE workspace built, and that refusal is NOT MEASURED, not 'not applicable to me'. I did not build the full workspace for a one-line string edit. Mitigating evidence, not a substitute: the package's own tsc --noEmit is green, and a string literal inside an existing test file already in that tsc program cannot move a per-entry count. Left to CI; I did not move any ratchet baseline.",
        "Repo-wide eslint (pnpm lint) — not run locally; CI owns the whole-farm run. I am NOT claiming a proven narrowing here, so record this as 'not run' rather than as a measured narrowing.",
        "The suite on UNMODIFIED main — I measured 313 passing on my branch, but never ran main's baseline, so 'zero assertions moved' rests on the diff being a single string literal (1 file, 1 insertion, 1 deletion, inside an it() title), not on a before/after count comparison. Nothing else moved in the diff."
      ],
      "open_questions": [],
      "other_v11_occurrences_found_and_left_alone": [
        "packages/plugins/plugin-audit/CHANGELOG.md:287 — '`session.tenantId` alias (#3280) was removed repo-wide in the v11 major'. Same factual claim as the test title I fixed, and wrong the same way, BUT it is a published append-only changelog entry and fence 1 of the dispatch plus the #9734 maintainer ruling (2026-08-22) put it out of scope. NOT fixed, NOT filed — the dispatch reserved that call for the PM.",
        "packages/plugins/plugin-audit/CHANGELOG.md:387 — 'hook/action session surface in v11 (#3290)'. Same class, same fence, same disposition.",
        "Those two are the ONLY other v11 occurrences anywhere in packages/plugins/plugin-audit (grep -rn 'v11' over the whole package). No source file besides the one I fixed carried the wrong label."
      ],
      "out_of_scope_findings": [],
      "escalate": [
        "The two plugin-audit CHANGELOG.md v11 sites above need a PM decision: adjudicate them onto #9872 (devx) or a new card, or rule them correct-as-history and close the question. I deliberately filed no issue for them because dispatch fence 5 said to report rather than file.",
        "No collision with PR #11522 (card #11424) materialized — it touches src/objects/sys-activity*.ts in the same package, my diff is src/audit-writers.test.ts. Branched from origin/main at daacc1071c; nothing to resolve.",
        "Shared verify-lock contention is real but did not block delivery: waits of 296s, 288s, 134s, 15s, 8s across five acquisitions, one holder observed at 318s. Consistent with #11363; no queue-timeout, no exit 99, no unexplained 143."
      ]
    }

    Generated by Claude Code

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    domain:servicesRetired 2026-10-22, alias of domain:server (services surface); in-flight cards keep it until landedtests

    Type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions