test(e2e): add Playwright spec for /play Bevy canvas route

play_canvas.spec.js covers the window.__FERROUS_DEBUG__ bridge on the /play route (five tests): bridge availability + seed param, draw3 URL param, applyLegalMove/undo round-trip, failureReport schema, and autonomous autoplay invariant batch across 7 seeds. All tests drive exclusively through the debug bridge — no DOM selectors, because the Bevy canvas is a single <canvas> element with no HTML controls. Also update SESSION_HANDOFF.md to reflect post-v0.35.1 work (10 commits since 2026-05-18 handoff), new e2e architecture notes, and HiDPI fix doc. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-02 14:03:25 -07:00
parent 2cf728210e
commit 2b1ad2161a
2 changed files with 130 additions and 6 deletions
@@ -0,0 +1,81 @@
+const { test, expect } = require("@playwright/test");
+
+async function gotoReadyPlay(page, seed = 42, draw3 = false) {
+    const suffix = draw3 ? "&draw3=" : "";
+    await page.goto(`/play?seed=${seed}${suffix}`);
+    await page.waitForFunction(
+        () =>
+            typeof window.__FERROUS_DEBUG__ === "object" &&
+            window.__FERROUS_DEBUG__.seed() !== null,
+        null,
+        { timeout: 30_000 }
+    );
+}
+
+test("play loads and exposes debug bridge", async ({ page }) => {
+    await gotoReadyPlay(page, 42);
+
+    const seed = await page.evaluate(() => window.__FERROUS_DEBUG__.seed());
+    expect(seed).toBe(42);
+
+    const legalMoves = await page.evaluate(() => window.__FERROUS_DEBUG__.legalMoves());
+    expect(Array.isArray(legalMoves)).toBeTruthy();
+    expect(legalMoves.length).toBeGreaterThan(0);
+});
+
+test("play respects draw3 URL param", async ({ page }) => {
+    await gotoReadyPlay(page, 77, true);
+
+    const snap = await page.evaluate(() => window.__FERROUS_DEBUG__.snapshot());
+    expect(snap).not.toBeNull();
+    expect(snap.draw_mode).toBe("DrawThree");
+});
+
+test("play debug bridge apply and undo work", async ({ page }) => {
+    await gotoReadyPlay(page, 42);
+
+    const baseline = await page.evaluate(() => window.__FERROUS_DEBUG__.moveHistory().length);
+
+    const applied = await page.evaluate(() => window.__FERROUS_DEBUG__.applyLegalMove(0));
+    expect(applied?.ok).toBeTruthy();
+
+    await expect
+        .poll(() => page.evaluate(() => window.__FERROUS_DEBUG__.moveHistory().length))
+        .toBe(baseline + 1);
+
+    const undone = await page.evaluate(() => window.__FERROUS_DEBUG__.undo());
+    expect(undone?.ok).toBeTruthy();
+
+    await expect
+        .poll(() => page.evaluate(() => window.__FERROUS_DEBUG__.moveHistory().length))
+        .toBe(baseline);
+});
+
+test("play failure report contains replay diagnostics", async ({ page }) => {
+    await gotoReadyPlay(page, 42);
+
+    const report = await page.evaluate(() => window.__FERROUS_DEBUG__.failureReport());
+    expect(report).not.toBeNull();
+    expect(typeof report.seed).toBe("number");
+    expect(Array.isArray(report.moveHistory)).toBeTruthy();
+    expect(Array.isArray(report.legalMoves)).toBeTruthy();
+    expect(report.currentState).toBeTruthy();
+    expect(report.invariants).toBeTruthy();
+});
+
+test("play autonomous autoplay keeps invariants stable across seed batch", async ({ page }) => {
+    test.setTimeout(120_000);
+    const seeds = [0, 1, 2, 5, 13, 42, 77];
+
+    for (const seed of seeds) {
+        await gotoReadyPlay(page, seed);
+        const run = await page.evaluate(() =>
+            window.__FERROUS_DEBUG__.runAutoplay({
+                maxSteps: 220,
+                maxVisitsPerState: 2,
+                policy: "loop_aware",
+            })
+        );
+        expect(run.ok, `seed ${seed} failed: ${JSON.stringify(run)}`).toBeTruthy();
+    }
+});