test(streams): regression tests for review bugs + Stream DO test harness by jonastemplestein · Pull Request #1455 · iterate/iterate

jonastemplestein · 2026-06-10T14:04:11Z

What

Stage 0 of an in-depth adversarial review of packages/streams (full plan + all findings in tasks/streams-review-fixes.md). This PR is tests only — no behavior changes. It pins the bugs the review found and gives the package its first always-on Durable Object tests, so the follow-up fix PRs land on a safety net.

Why first

The ~1000-line Stream DO had zero always-on tests (append, idempotency, the pause gate, >512KB chunking, replay all lived only in example-app/e2e/*, skipped unless STREAM_STAGING_E2E=true). The review surfaced two CRITICAL data-loss/security bugs and several MAJORs; this PR documents them as failing tests before any fix touches the code.

The bug-pins (committed as `it.fails` ratchets)

The six tests below assert the correct behavior, which is currently violated — so they register as "expected fail" and keep pnpm test green. Each was proven to fail for the right reason (output in the table). When a fix lands, flip it.fails → it.

Test	Finding	Proven failure today
T1	C1 swallowed-ingest desync (CRITICAL)	`{total: 7}` — event lost, expected `12`
T2	C2 writeState checkpoint ordering (CRITICAL)	nothing persisted on retry
T3	C3 `PublicStreamRpcTarget` leaks `protected` methods (CRITICAL, security)	prototype exposes `writeCoreProcessorState`
T4	M2 `source` field crashes append	`Unrecognized key: "source"`
T5	M3 breaker token bucket on backwards clock	`-99996` tokens
T6	M4 breaker silent after replay trip	`0` pauses

T7 (idempotency) and T8 (DO smoke: offsets, getEvents paging, >512KB chunk round-trip, subscribe replay) are coverage for already-correct behavior and pass now.

The Stream DO test harness (structural)

Adds vitest-pool-workers to packages/streams (mirroring packages/shared):

vitest.workers.config.ts + vitest.workers.jsonc (binds the Stream + StreamProcessorRunner DOs)
src/workers/test-entry.ts (minimal worker exporting the DOs)
node config excludes **/*.workers.test.ts; pnpm test runs both pools (test:node / test:workers)

Following cloudflare/agents and the workers-sdk rpc fixture, the DO tests use runInDurableObject(stub, (instance, state) => …) to call the instance directly and inspect state.storage — clean local throws (no RPC unhandled-rejection workaround), deterministic replay, and the chunking test asserts the event really spans multiple event_chunks rows. One test stays on the RPC stub to cover the production boundary.

Verification

node pool:    58 passed | 4 expected-fail
workers pool:  6 passed | 2 expected-fail
typecheck: clean   lint: clean

🤖 Generated with Claude Code

Note

Low Risk
Test-only changes and CI script updates; no production paths modified. Bug-pin tests document known failures without changing runtime behavior.

Overview
Stage 0 (tests only) for the packages/streams review: no runtime behavior changes. It adds the first always-on coverage for the Stream Durable Object and pins six known bugs as it.fails ratchets so pnpm test stays green until fixes land.

A vitest-pool-workers harness is introduced (vitest.workers.config.ts, vitest.workers.jsonc, src/workers/test-entry.ts), with @cloudflare/vitest-pool-workers as a dev dependency. Node Vitest excludes **/*.workers.test.ts; pnpm test runs both pools (test:node / test:workers).

Node (stream-review-regressions.test.ts): T1–T2 (ingest desync, checkpoint after writeState failure) and T5–T6 (circuit-breaker clock regression and post-replay flood) assert correct behavior via stubs/fakes.

Workerd (stream.workers.test.ts): T3–T4 are failing ratchets for RPC surface leak and source on append; T7–T8 pass now (idempotency, offsets, paging, >512KB chunking, subscribe replay), mostly via runInDurableObject with one RPC-stub smoke test.

tasks/streams-review-fixes.md marks Stage 0 complete and documents the harness pattern for later fix PRs.

^{Reviewed by Cursor Bugbot for commit 5c94b0e. Bugbot is set up for automated code reviews on this repo. Configure here.}

Stage 0 of the packages/streams review (tasks/streams-review-fixes.md). Pins the bugs the review found and gives the package its first always-on Durable Object tests. Node tests (stream-review-regressions.test.ts), proven red: - T1 (C1): a swallowed ingest failure permanently drops events — drives the real createStreamProcessorHost with a fake pump that mimics the Stream DO delivery path (cursor advances before delivery, result fire-and-forgotten). - T2 (C2): a writeState failure advances the in-memory checkpoint, so the retried batch persists nothing. - T5 (M3): the circuit-breaker token bucket drains on a backwards clock. - T6 (M4): the breaker stays silent on a live flood after tripping during replay. vitest-pool-workers harness (the structural gap — the package had zero always-on DO tests): vitest.workers.config.ts + vitest.workers.jsonc + src/workers/test-entry.ts. Node config excludes *.workers.test.ts; `pnpm test` runs both pools. Stream DO tests (stream.workers.test.ts): - T3 (C3, red): PublicStreamRpcTarget proxies protected readCoreProcessorState / writeCoreProcessorState. - T4 (M2, red): appending an event with a `source` field throws. - T7, T8 (green coverage): idempotency-key dedupe + offset precondition, consecutive offsets, getEvents paging, >512KB chunk round-trip (asserted via direct event_chunks inspection), and subscribe replay. The six bug-pins are committed as `it.fails` ratchets so `pnpm test` stays green until the fixes land (flip `it.fails` -> `it` then). Following cloudflare/agents and the workers-sdk rpc fixture, the DO tests use runInDurableObject for direct instance + storage access (clean local throws, deterministic replay, direct SQL inspection), keeping one stub-RPC test for the production boundary. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

… surface (C1/C2/C3) (#1458) ## What **Stage 1** of the `packages/streams` review (plan in `tasks/streams-review-fixes.md`). Fixes the three CRITICAL findings. Flips the Stage 0 ratchets **T1, T1b, T2, T3** from `it.fails` → `it`. **Stacked on #1455** (Stage 0 tests) — base is `streams-review-stage0-tests`; retarget to `main` once that merges. ## C1 — a failed batch no longer drops events (data loss) The delivery pump is fire-and-forget and advances its cursor before delivery, so a swallowed ingest failure left a permanent gap in the subscriber's state. Per the agreed approach (consumer resubscribes from checkpoint): - On ingest failure the host re-handshakes from the durable checkpoint; the stream replays the batch (replay is idempotent). - **Continued-delivery race:** because the pump keeps delivering during recovery, a *later* batch's successful ingest would advance the checkpoint past the gap. Each subscription is now tagged with a `generation`, and ingest runs through a per-processor serial chain that re-checks the generation **between** batches — batches from the superseded connection are dropped, and the post-recovery replay is the single source of truth. - **Poison policy (owner's call):** after `MAX_CONSECUTIVE_INGEST_FAILURES` (3), the host appends a `stream/error-occurred` event and disconnects rather than hot-looping; recovery is then up to the subscriber/processor. Covered by **T1** (transient recovery under continued delivery) + **T1b** (poison → error + disconnect). > Note: the browser-mirror consumer (`stream-browser-store.ts`) has the same swallow-and-wedge bug. Left for Stage 4 with the other browser-runtime fixes (tracked in the task). ## C2 — checkpoint ordering (silent data loss on write failure) `StreamProcessor.#ingest` advanced `#state`/`#checkpointOffset` *before* `writeState`, so a failed durable write left the in-memory checkpoint ahead and the retried batch silently no-oped. Now it persists the snapshot **first**, then advances. (Inlined the single-use `#saveSnapshot`.) Covered by **T2**. ## C3 — `PublicStreamRpcTarget` method leak (security) `makeRpcTargetClass` copied every prototype method except a two-item denylist, so the `protected` `readCoreProcessorState` / `writeCoreProcessorState` were proxied onto the unauthenticated public target — `writeCoreProcessorState` could inject an attacker-chosen subscription callable that the DO later dispatches with full `env`. Added an `include` allowlist; both Stream targets now allowlist the explicit `StreamRpc` surface (`satisfies readonly (keyof StreamRpc)[]` so it can't drift). Covered by **T3**. ## Verification ``` node pool: 61 passed | 2 expected-fail (T5/T6 remain → Stage 3) workers pool: 7 passed | 1 expected-fail (T4 remains → Stage 3) typecheck: clean lint: clean ``` 🤖 Generated with [Claude Code](https://claude.com/claude-code)  --- > [!NOTE] > **High Risk** > Changes durable ingest/checkpoint behavior and closes a security hole on unauthenticated Stream RPC; incorrect recovery or ordering could still cause data loss or wedge delivery. > > **Overview** > **Stage 1** closes three CRITICAL streams review findings (C1/C2/C3) and turns regression tests T1, T1b, T1c, T2, and T3 from `it.fails` into passing `it`. > > **C1 — failed batches no longer lose events (hosted path).** `createStreamProcessorHost` no longer swallows ingest errors. On failure it re-handshakes from the durable checkpoint so the stream replays the batch. A **generation** gate plus a per-processor **ingest chain** drops batches from superseded connections while recovery runs (avoids advancing the checkpoint past a gap). After three consecutive failures it appends `stream/error-occurred` and disconnects instead of hot-looping. Regression coverage adds a replay-capable `fakeStream`, T1b (poison), and T1c (recovery throw must not wedge the chain). > > **C2 — checkpoint vs durable write.** `StreamProcessor.#ingest` now **`await`s `writeState` before** updating in-memory `#state` / `#checkpointOffset`, so a failed write leaves the batch retryable instead of a silent no-op. > > **C3 — public RPC surface.** `makeRpcTargetClass` gains an **`include` allowlist**; `PublicStreamRpcTarget` / `StreamRpcTarget` only expose explicit `StreamRpc` methods so protected `readCoreProcessorState` / `writeCoreProcessorState` are not callable on the unauthenticated public target. > > Task doc and test headers are updated to mark C1/C2/C3 fixed on the hosted path; browser mirror recovery remains Stage 4. > > Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit e5aeaac. Bugbot is set up for automated code reviews on this repo. Configure [here](https://www.cursor.com/dashboard/bugbot).   ## Environment Config Lease    Lease: `preview-4` Doppler config: `preview_4` Type: `environment-config-lease` Leased until: 2026-06-10T16:10:58.241Z ### OS Status: tests failed Commit: `a9d4a4d` Preview: https://os.iterate-preview-4.com Summary: ❯ node src/itx/e2e/itx.e2e.test.ts (10 tests | 1 failed | 1 skipped) 93401ms [Workflow run](https://github.com/iterate/iterate/actions/runs/27285949769) Updated: 2026-06-10T15:23:10.059Z <details> <summary>Failure details</summary> <pre>...(truncated) t.toBeNull(); 397| expect(error!.name).toBe("ItxError"); | ^ 398| expect(error!.code).toBe("NOT_FOUND"); 399| expect(error!.details).toEqual({ projectIdOrSlug: "definitely-not-a-… ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯[1/1]⎯ > @iterate-com/os@0.0.1 e2e /home/runner/work/iterate/iterate/apps/os > vitest --config e2e/vitest.config.ts -t 'OS preview smoke' [vitest-artifacts] run root: /tmp/os-e2e-deTncv [vitest] run slug: os-vitest-run-20260610-151920 RUN v4.0.15 /home/runner/work/iterate/iterate/apps/os stdout | e2e/vitest/preview-smoke.e2e.test.ts > OS preview smoke OS preview smoke passed for https://os.iterate-preview-4.com/ ✓ e2e/vitest/preview-smoke.e2e.test.ts (1 test) 6669ms ✓ OS preview smoke 6668ms ↓ e2e/vitest/admin-project.e2e.test.ts (1 test | 1 skipped) ↓ e2e/vitest/agents.e2e.test.ts (9 tests | 9 skipped) ↓ e2e/vitest/codemode-mcp-provider-stack.e2e.test.ts (1 test | 1 skipped) ↓ e2e/vitest/codemode.e2e.test.ts (3 tests | 3 skipped) ↓ e2e/vitest/e2e-test-map.e2e.test.ts (13 tests | 13 skipped) Test Files 1 passed | 5 skipped (6) Tests 1 passed | 27 skipped (28) Start at 15:19:20 Duration 7.99s (transform 210ms, setup 0ms, import 837ms, tests 6.67s, environment 1ms) > @iterate-com/os@0.0.1 e2e:itx /home/runner/work/iterate/iterate/apps/os > vitest run --config src/itx/e2e/vitest.config.ts --project node [vitest-artifacts] run root: /tmp/os-itx-e2e-DurFwM [vitest] run slug: os-vitest-run-20260610-151929 RUN v4.0.15 /home/runner/work/iterate/iterate/apps/os ❯ node src/itx/e2e/itx.e2e.test.ts (10 tests | 1 failed | 1 skipped) 93401ms ✓ itx scripts run identically over Cap'n Web and /api/itx/run 11957ms ✓ the five-step capability flow: provide live, call, promote durable, call from a script 11979ms ✓ platform bindings are dialable capabilities (raw + wrapped) 11831ms ↓ the first-party McpClient cap bridges a remote MCP server ✓ script executions leave a two-event record on the /itx stream 13369ms ✓ worker caps hold a correctly scoped itx of their own 9833ms ✓ members caps auto-proxy every public method/getter at any depth 11350ms ✓ one dynamic worker cap calls another's methods through its own itx 10597ms × kernel errors cross capnweb as ItxError-shaped errors with codes 89ms ✓ revoked and offline caps fail with instructive errors 10399ms ✓ node src/itx/e2e/itx-fork.e2e.test.ts (3 tests) 52757ms ✓ fork: child caps shadow the parent, misses delegate up the chain 15585ms ✓ fork narrows access: a session cannot reach sibling projects 19542ms ✓ fork: child worker caps run with the owning project's authority 15835ms ✓ node src/itx/e2e/itx-egress.e2e.test.ts (3 tests) 37217ms ✓ itx.fetch substitutes secrets through project egress (explicit door) 9642ms ✓ bare fetch() in a project itx script goes through egress (implicit door) 16508ms ✓ bare fetch() inside a worker cap goes through egress (implicit door) 8802ms ✓ node src/itx/e2e/itx-http.e2e.test.ts (2 tests) 24080ms ✓ facet caps keep private durable state across invocations 8799ms ✓ HTTP-exposed caps serve their own hostname: admin, share URL, public 13750ms ✓ node src/itx/e2e/itx-subscribe.e2e.test.ts (1 test) 12735ms ✓ subscribe replays history, tails live appends, and unsubscribes 10991ms Test Files 1 failed | 4 passed (5) Tests 1 failed | 17 passed | 1 skipped (19) Start at 15:19:29 Duration 220.79s (transform 79ms, setup 0ms, import 184ms, tests 220.19s, environment 0ms) ::error file=/home/runner/work/iterate/iterate/apps/os/src/itx/e2e/itx.e2e.test.ts,title=[node] src/itx/e2e/itx.e2e.test.ts > kernel errors cross capnweb as ItxError-shaped errors with codes,line=397,column=23::AssertionError: expected 'Error' to be 'ItxError' // Object.is equality%0A%0AExpected: "ItxError"%0AReceived: "Error"%0A%0A ❯ src/itx/e2e/itx.e2e.test.ts:397:23%0A%0A ELIFECYCLE Command failed with exit code 1.</pre> </details>  --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

#1479) ## Streams review — Stages 4–7 (browser robustness, perf, dead code, docs) The final batch of the `packages/streams` adversarial review (`tasks/streams-review-fixes.md`), bundled into one PR per request. Stages 0/1/3 already shipped (#1455, #1458, #1459); **Stage 2 (lazy init) was abandoned** — #1460's subscriber-presence model appends a `subscriber-connected` fact on every `subscribe()`, which fundamentally conflicts with "don't initialize storage until the first `append()`". `created`/`woken` stay eager. All findings were re-verified against current `main` after #1460/#1457 moved the codebase; items those PRs already fixed or made obsolete are called out below. ### Stage 4 — browser-runtime correctness - **C1-browser** — the server delivery pump is fire-and-forget for inbound subscribers and never reports a failed `ingest`, so a browser mirror that fails to apply a batch silently desyncs forever. The browser now self-heals: `ingestWithSelfHeal` resubscribes from the persisted checkpoint with bounded exponential backoff on ingest failure. - **B1** — connection-epoch guard so a stale connection's late status callback can't clobber the live connection. - **B2** — `appendBatch`/`runtimeState` await connection readiness (`whenStreamReady`) instead of throwing "disposed" during a transient reconnect; only throw when actually disposed. - **B3** — the `ROLLBACK` in `stream-db.worker.ts` `batch()` is now in its own try/catch so it can't mask the original error (which `withBusyRetry`/`isBusyError` need to see). - **B4** — reconcile guards on stream incarnation (server `createdAt` + a `mirror_meta` table): rebuild the mirror on incarnation change rather than trusting the offset comparison after a `reset()`. - **B5** — `AbortSignal` on the Web Lock request (released on disposal) + the request rejection is surfaced instead of `void`-swallowed. - **B6** — query lifecycle: arm GC at query creation (not only on unsubscribe), skip listenerless queries in `#onChange`, and equality-check before notifying to avoid spurious `useSyncExternalStore` churn. ### Stage 5 — performance - **P1** — `browser-event-feed` O(n²) write amplification: coalesce to one op per `local_index` (was a cumulative op per event, each re-serializing the whole accumulated array). - **P2** — bound group rows with `MAX_GROUP_EVENTS = 200` so one dominant event type can't grow a single blob unboundedly. - **P3** — drop the per-event full-state `stateSchema.parse` in core `reduce`; state is validated only at the KV/recovery trust boundary now. - **P4** — already fixed by #1460 (subscribe override forwards `eventTypes`; filtering is server-side). No change. ### Stage 6 — dead code / elegance (~247 lines deleted) - **E1** dead exports in `shared/stream-processors.ts`; **E2** `waitForOpen`; **E3** unreachable circuit-breaker `paused` guard; **E5** dead `waitForEvent`/`messageInbox.error` machinery in `subscription.ts`; **E6** redundant circuit-breaker `consumes` entries. - **E4 obsolete** (#1460 removed `processor-registered`; the shared `circuit-breaker-types` is a live, separately-imported hierarchy, not dead code). **E7 declined** — post-#1460 the subscribe override genuinely needs to retain the client callback / wire `onRpcBroken`, so folding it into the generic `makeRpcTargetClass` isn't worth it. ### Stage 7 — docs - **D1** `design.md`: status banner, fixed 1-based offsets + core-event examples, the callable subscriber spec, SQLite/512KB-chunking storage, live-tail `replayAfterOffset` default, OPFS mirror; superseded banners on the never-shipped `implementProcessor`/`connectStream` sections. - **D2** `README.md`: real browser export (`withStreamConnectionFromBrowser`), sync `Disposable`, `?path=` route map; new "Append & subscription semantics" section (D5). - **D3** ADR 0001 superseded with the shipped per-`(namespace, path, slug)` singleton model. - **D4** remaining comment drift (`beforeAppend`→`validateAppend`, `afterAppendBatch`→`processEventBatch`). ### Testing `pnpm typecheck`, `pnpm lint`, node (79) + workers (15) suites, and example-app typecheck all green. Example-app Playwright e2e run locally against a clean `vite dev` (real Stream DO via Miniflare): **26 passed**. `origin/main` merged in (no `packages/streams` overlap; the new project-config stream-processor consumers don't touch any deleted exports). 🤖 Generated with [Claude Code](https://claude.com/claude-code)  --- > [!NOTE] > **Medium Risk** > Touches browser reconnect, mirror discard, and ingest self-heal paths that affect live viewers and local SQLite mirrors; server stream DO behavior is mostly unchanged aside from core reduce hot-path optimization. > > **Overview** > Closes out **Stages 4–7** of the `packages/streams` adversarial review: browser mirror reliability, feed/core hot-path perf, dead-code removal, and doc alignment. Stage 2 (lazy stream init) stays **abandoned** and is only noted in `tasks/streams-review-fixes.md`. > > **Browser runtime** (`stream-browser-store.ts` and friends) is reworked so the OPFS mirror can recover instead of silently drifting. Inbound delivery is fire-and-forget, so failed `ingest` now triggers **`ingestWithSelfHeal`** (resubscribe from the last checkpoint with capped exponential backoff). Reconnects go through one **`scheduleReconnect`** path with a **connection epoch** so stale WebSocket callbacks cannot tear down the replacement connection. **`appendBatch` / `runtimeState`** use **`callWhenReady`** / **`whenStreamReady`** during transient reconnects instead of throwing “disposed”. Mirror trust uses server **`createdAt`** as incarnation identity via new **`mirror_meta`** helpers in **`stream-browser-db.ts`**. Web Lock **`release()`** aborts pending lock requests; the SQLite worker preserves the original batch error if `ROLLBACK` fails; reactive queries GC earlier, skip unobserved refreshes, and avoid notify when snapshots are unchanged. > > **Performance:** `browser-event-feed` **`planFeedOps`** coalesces one SQL op per group row and caps groups at **`MAX_GROUP_EVENTS` (200)**. Core **`reduce`** no longer exit-parses the full `stateSchema` on every appended event. > > **Cleanup:** ~247 lines removed (unused `stream-processors` exports, `waitForOpen`, dead subscription `waitForEvent` machinery, trimmed circuit-breaker `consumes` and an unreachable guard). **Docs:** `README` (real browser API, `?path=` routes, append/subscription semantics), `design.md` and ADR 0001 marked superseded where they diverge from shipped code. > > Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit f52f305. Bugbot is set up for automated code reviews on this repo. Configure [here](https://www.cursor.com/dashboard/bugbot).   ## Environment Config Lease    Lease: `preview-3` Doppler config: `preview_3` Type: `environment-config-lease` Leased until: 2026-06-10T22:59:42.486Z ### OS Status: deployed Commit: `f52f305` Preview: https://os.iterate-preview-3.com [Workflow run](https://github.com/iterate/iterate/actions/runs/27309066809) Updated: 2026-06-10T22:02:58.860Z  --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

The packages/streams adversarial review is fully shipped — Stages 0/1/3 (#1455/#1458/#1459), Stages 4–7 (#1479), Stage 2 deliberately abandoned. This deletes the now-stale `tasks/streams-review-fixes.md` and removes the dangling pointers to it from the regression-test comments (the finding ids and the PRs in git history are the durable record). Doc/comment-only; no code changes. 🤖 Generated with [Claude Code](https://claude.com/claude-code)  --- > [!NOTE] > **Low Risk** > Documentation and comment-only changes; no runtime or test behavior changes. > > **Overview** > Removes the completed **`tasks/streams-review-fixes.md`** tracking doc now that the June 2026 `packages/streams` review work is shipped (Stages 0/1/3 via #1455/#1458/#1459, Stages 4–7 via #1479; Stage 2 abandoned). > > Regression test file headers are updated so they no longer link to that task file. Comments now cite **review finding ids (C*/M*)** and the **merged PRs** as the durable record instead. > > Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit 7c33b86. Bugbot is set up for automated code reviews on this repo. Configure [here](https://www.cursor.com/dashboard/bugbot).   ## Environment Config Lease    Lease: `preview-2` Doppler config: `preview_2` Type: `environment-config-lease` Leased until: 2026-06-11T10:34:11.088Z ### OS Status: deployed Commit: `7c33b86` Preview: https://os.iterate-preview-2.com [Workflow run](https://github.com/iterate/iterate/actions/runs/27337613329) Updated: 2026-06-11T09:38:48.453Z  Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

jonastemplestein and others added 3 commits June 10, 2026 15:03

[autofix.ci] apply automated fixes

46ad44d

[autofix.ci] apply automated fixes (attempt 2/3)

5c94b0e

jonastemplestein mentioned this pull request Jun 10, 2026

fix(streams): Stage 1 — failed-batch desync, checkpoint ordering, RPC surface (C1/C2/C3) #1458

Merged

jonastemplestein merged commit 6257b82 into main Jun 10, 2026
11 checks passed

jonastemplestein deleted the streams-review-stage0-tests branch June 10, 2026 15:00

jonastemplestein mentioned this pull request Jun 10, 2026

streams review: Stages 4–7 (browser robustness, perf, dead code, docs) #1479

Merged

jonastemplestein mentioned this pull request Jun 11, 2026

chore(streams): remove completed review task record #1491

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

test(streams): regression tests for review bugs + Stream DO test harness#1455

test(streams): regression tests for review bugs + Stream DO test harness#1455
jonastemplestein merged 3 commits into
mainfrom
streams-review-stage0-tests

jonastemplestein commented Jun 10, 2026 •

edited by cursor Bot

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

jonastemplestein commented Jun 10, 2026 • edited by cursor Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What

Why first

The bug-pins (committed as it.fails ratchets)

The Stream DO test harness (structural)

Verification

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

jonastemplestein commented Jun 10, 2026 •

edited by cursor Bot

Loading

The bug-pins (committed as `it.fails` ratchets)