test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
// Browser bundle entry for the V2 "items" collab e2e — the PRODUCTION stack
|
|
|
|
|
// (ysync 0008): startKicadCollab → connectKicadDoc + bindKicadCollab over
|
|
|
|
|
// moduleItemsBridge, Y keys kdoc_*, C++ exports kicadCollabSnapshotItems /
|
|
|
|
|
// kicadCollabApplyItems / window.kicadCollab.onItems.
|
|
|
|
|
//
|
|
|
|
|
// The sibling browser-entry.ts drives the LEGACY scalar wire (startCollab /
|
|
|
|
|
// onDelta), which is DEAD in production — nothing registers onDelta; WasmTool
|
|
|
|
|
// binds onItems only (ysync-review miss 11). New collab e2e must load THIS
|
|
|
|
|
// bundle; the legacy one stays only until the scalar wire is deleted.
|
|
|
|
|
//
|
|
|
|
|
// Build: npm run build:collab (tests/) → tests/apps/kicad/collab-bundle-v2.js
|
|
|
|
|
//
|
|
|
|
|
// IMPORTANT (build.mjs): yjs is aliased to ONE physical copy. web/standalone and
|
|
|
|
|
// web/pcbjam-shared are separate pnpm workspaces, so without the alias the
|
|
|
|
|
// bundle carries two yjs instances (the legacy bundle demonstrably does) — and
|
|
|
|
|
// the v2 path breaks on that (Y types are instanceof-checked singletons; same
|
|
|
|
|
// reason the standalone vitest config sets `dedupe: ["yjs"]`).
|
|
|
|
|
import * as Y from "yjs";
|
|
|
|
|
import {
|
2026-07-21 09:16:38 +02:00
|
|
|
compareSlots,
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
docToFile,
|
|
|
|
|
docToY,
|
2026-07-21 09:16:38 +02:00
|
|
|
driftDocDelta,
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
fileToDoc,
|
|
|
|
|
isEmptyKicadDelta,
|
|
|
|
|
yToDoc,
|
|
|
|
|
} from "@pcbjam/shared";
|
|
|
|
|
import {
|
|
|
|
|
attachKicadCollab,
|
|
|
|
|
connectKicadDoc,
|
|
|
|
|
type KicadCollabHandle,
|
|
|
|
|
type KicadItemsModule,
|
|
|
|
|
type KicadItemsWindow,
|
|
|
|
|
} from "../../web/standalone/src/wasm/collab/index";
|
|
|
|
|
|
|
|
|
|
interface StartOpts {
|
|
|
|
|
room: string;
|
|
|
|
|
/** BroadcastChannel settle window before the seed-vs-adopt decision. */
|
|
|
|
|
settleMs?: number;
|
|
|
|
|
/**
|
|
|
|
|
* Full file text; when set and the room is EMPTY, the Y.Doc is file-seeded
|
|
|
|
|
* from fileToDoc(seedText) — the production first-tab path (and the branch
|
|
|
|
|
* bug 01 lives in). Omit to exercise the editor-snapshot / adopt branches.
|
|
|
|
|
*/
|
|
|
|
|
seedText?: string;
|
|
|
|
|
/**
|
|
|
|
|
* The ydoc-load path: the editor opened exactly this doc's content, so seed
|
|
|
|
|
* only baselines the wasm differ instead of running the adopt apply.
|
|
|
|
|
*/
|
|
|
|
|
editorMatchesDoc?: boolean;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Start the v2 stack; the handle lands on window.__collabV2 for in-page asserts. */
|
|
|
|
|
async function start(
|
|
|
|
|
mod: KicadItemsModule,
|
|
|
|
|
win: KicadItemsWindow,
|
|
|
|
|
opts: StartOpts,
|
|
|
|
|
): Promise<void> {
|
|
|
|
|
// startKicadCollab's body, split so editorMatchesDoc is reachable (the
|
|
|
|
|
// production WasmTool uses the same connect + attach pair for ydoc mode).
|
|
|
|
|
const session = await connectKicadDoc({
|
|
|
|
|
provider: { kind: "broadcastchannel", settleMs: opts.settleMs ?? 400 },
|
|
|
|
|
room: opts.room,
|
|
|
|
|
});
|
|
|
|
|
const h = attachKicadCollab(mod, win, session, {
|
|
|
|
|
seedDoc: opts.seedText ? fileToDoc(opts.seedText) : undefined,
|
|
|
|
|
editorMatchesDoc: opts.editorMatchesDoc,
|
|
|
|
|
});
|
|
|
|
|
(window as unknown as { __collabV2?: KicadCollabHandle }).__collabV2 = h;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
function handle(): KicadCollabHandle {
|
|
|
|
|
const h = (window as unknown as { __collabV2?: KicadCollabHandle }).__collabV2;
|
|
|
|
|
if (!h) throw new Error("KicadCollabV2: start() has not completed");
|
|
|
|
|
return h;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** docToFile of the live room doc — THROWS if the doc stopped materializing. */
|
|
|
|
|
function renderActiveDoc(): string {
|
|
|
|
|
return docToFile(yToDoc(handle().doc));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** What ONE seeder would materialize — the bug-06 reference rendering. */
|
|
|
|
|
function singleSeedRender(seedText: string): string {
|
|
|
|
|
const ydoc = new Y.Doc();
|
|
|
|
|
try {
|
|
|
|
|
docToY(fileToDoc(seedText), ydoc);
|
|
|
|
|
return docToFile(yToDoc(ydoc));
|
|
|
|
|
} finally {
|
|
|
|
|
ydoc.destroy();
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
export interface DriftSummary {
|
|
|
|
|
added: string[];
|
|
|
|
|
updated: string[];
|
|
|
|
|
removed: string[];
|
2026-07-21 09:16:38 +02:00
|
|
|
/** Order-only item churn — surfaced for triage, NOT drift (kicad-delta.ts). */
|
|
|
|
|
reordered: string[];
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
layoutChanged: boolean;
|
2026-07-21 09:16:38 +02:00
|
|
|
/** Order-only layout churn — likewise not report-worthy on its own. */
|
|
|
|
|
layoutReordered: boolean;
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
metaChanged: boolean;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* The drift-detect convergence oracle, replicating computeDrift's core
|
2026-07-21 09:16:38 +02:00
|
|
|
* (web/standalone/src/wasm/collab/drift-detect.ts) from @pcbjam/shared
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
* primitives only — drift-detect itself pulls `@/lib/api`, so it can't be
|
|
|
|
|
* bundled here. Serializes the live model via the tool's save fn, diffs it
|
2026-07-21 09:16:38 +02:00
|
|
|
* against the room doc with the PRODUCTION comparator (driftDocDelta +
|
|
|
|
|
* compareSlots — order-only churn goes to reordered/layoutReordered and does
|
|
|
|
|
* not make a report, matching ysync 0010); null means editor ≡ doc.
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
*/
|
|
|
|
|
function driftReport(saveFn: string, scratchPath: string): DriftSummary | null {
|
|
|
|
|
const w = window as unknown as {
|
|
|
|
|
Module: Record<string, (p: string) => void>;
|
|
|
|
|
FS: {
|
|
|
|
|
readFile(p: string, o: { encoding: "utf8" }): string;
|
|
|
|
|
unlink(p: string): void;
|
|
|
|
|
};
|
|
|
|
|
};
|
|
|
|
|
w.Module[saveFn]!(scratchPath);
|
|
|
|
|
let text: string;
|
|
|
|
|
try {
|
|
|
|
|
text = w.FS.readFile(scratchPath, { encoding: "utf8" });
|
|
|
|
|
} finally {
|
|
|
|
|
try {
|
|
|
|
|
w.FS.unlink(scratchPath);
|
|
|
|
|
} catch {
|
|
|
|
|
/* scratch cleanup is best-effort */
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
const wasmDoc = fileToDoc(text);
|
|
|
|
|
const ydocDoc = yToDoc(handle().doc);
|
2026-07-21 09:16:38 +02:00
|
|
|
const diff = driftDocDelta(ydocDoc, wasmDoc);
|
|
|
|
|
const layoutRelation = compareSlots(ydocDoc.layout, wasmDoc.layout);
|
|
|
|
|
const layoutChanged = layoutRelation === "different";
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
const metaChanged = ydocDoc.root !== wasmDoc.root;
|
2026-07-21 09:16:38 +02:00
|
|
|
// layoutReordered deliberately outside the gate — same as computeDrift.
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
if (isEmptyKicadDelta(diff) && !layoutChanged && !metaChanged) return null;
|
|
|
|
|
return {
|
|
|
|
|
added: diff.added.map((i) => i.uuid),
|
|
|
|
|
updated: diff.updated.map((i) => i.uuid),
|
|
|
|
|
removed: diff.removed,
|
2026-07-21 09:16:38 +02:00
|
|
|
reordered: diff.reordered.map((i) => i.uuid),
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
layoutChanged,
|
2026-07-21 09:16:38 +02:00
|
|
|
layoutReordered: layoutRelation === "reordered",
|
test(ysync): repro tests for review bugs 01-07 + v2 items-wire e2e port (miss 11)
The 2026-07-02 sync review (docs/features/ysync-review, ysync-review branch)
found 7 bugs and that the two-tab e2e only exercised the DEAD legacy scalar
wire. This lands plan doc 15 in full; results + empirical findings in doc 16.
- tests/collab/browser-entry-v2.ts (+build.mjs): the PRODUCTION v2 stack
bundled for e2e (connectKicadDoc + attachKicadCollab, kdoc_* keys), with
in-page renderActiveDoc/singleSeedRender/driftReport helpers and yjs forced
to ONE copy (the two web pnpm workspaces otherwise bundle two
instanceof-incompatible instances).
- tests/kicad/ysync-two-tab.spec.ts: pl_editor green baseline (A↔B edits,
ITEM-level drift silence) + divergent-uuid adopt; bug-01 pcb/ee fresh-room
repros (Chromium-only: two kicad_editor tabs exceed Firefox's per-process
wasm budget); bug-06 concurrent-seed race; bug-03 Y-half.
- tests/kicad/ysync-repros-{pcbnew,eeschema}.spec.ts: bugs 02/03/05 + the
bug-04 matrix (anchor-centred fp rotation, pad resize, endpoint drag,
symbol rotation, Value-field edit), each with green landed-preconditions;
the "local move emits" controls double as headless-emit probes — GREEN on
both tools, so every emit-dependent repro is a live test.fail.
- wasm/bindings: 7 local-edit test hooks via real commits
(CallAfter+COROUTINE) — TestRemoveItem/TestRotateItem (both tools,
dispatched in the merged image), TestSetPadSize/TestMoveEndpoint (pcbnew),
TestSetFieldText (eeschema).
- web/standalone ysync-repros.test.ts: bug-01 units (C++-faithful fake gating
emit on ensureBridge) + bug-07a/b (stale DOWN hook, real sheet-manager gap).
- web/pcbjam-shared bump: bug-03/06 unit repros.
Convention: every repro asserts the CORRECT behavior and is expected-fail
(test.fail/it.fails) naming its bug doc; a fix flips it to "unexpected pass",
forcing marker removal — the repro becomes the regression test. Every
expected failure verified (JSON reporter) to fail at its documented assert.
Suite state: 39 passed / 0 failed / 0 flaky / 5 skipped (2 firefox guards,
2 pre-existing legacy two-tab skips, 1 pre-existing roundtrip fixme).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DPfrVhfYgPPgtawjSssZfn
2026-07-03 12:43:35 +02:00
|
|
|
metaChanged,
|
|
|
|
|
};
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
declare global {
|
|
|
|
|
interface Window {
|
|
|
|
|
KicadCollabV2?: {
|
|
|
|
|
start: typeof start;
|
|
|
|
|
renderActiveDoc: typeof renderActiveDoc;
|
|
|
|
|
singleSeedRender: typeof singleSeedRender;
|
|
|
|
|
driftReport: typeof driftReport;
|
|
|
|
|
};
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
window.KicadCollabV2 = { start, renderActiveDoc, singleSeedRender, driftReport };
|