feat(producer): seedable Math.random / crypto.getRandomValues shim, gated

Part of Phase 2 of the distributed rendering plan (determinism hardening). See DISTRIBUTED-RENDERING-PLAN.md §5.2 (Math.random row) and §17.2 (gating table). The existing `VIRTUAL_TIME_SHIM` freezes Date.now / performance.now / rAF on a render seek but leaves `Math.random` and `crypto.getRandomValues` as native non-deterministic. Compositions that paint stochastic visuals through these APIs produce different pixels on distributed retries. This change adds `buildVirtualTimeShim({ seedRandomFromFrame: boolean })`. Default `false` returns a string byte-identical to today's `VIRTUAL_TIME_SHIM` (pinned by a new unit test). When `true`, the script additionally: - Installs a Mulberry32 PRNG with a single uint32 state - Reseeds the state from the current virtual time on every `seekToTime(ms)` call (Knuth multiplicative hash + golden-ratio offset) - Replaces `Math.random` with the PRNG output - Replaces `crypto.getRandomValues` to fill the buffer from the PRNG `VIRTUAL_TIME_SHIM` (the const consumed by `renderOrchestrator` + `probeStage`) is now `buildVirtualTimeShim({ seedRandomFromFrame: false })` — in-process behavior unchanged, producer regression baselines unaffected. Phase 3 distributed primitives will pass `true` when building the chunk worker's file-server scripts. 10 new unit tests at packages/producer/src/services/ fileServer-seededRandom.test.ts use node:vm to evaluate the shim in isolated contexts and pin both branches: - default emits no RNG override and leaves Math.random native - locked emits the seeded block, produces identical sequences across fresh VMs at the same time, and yields different sequences for different times This is part of a stack of 10 PRs; this is PR 4 of 10. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-09-03 12:54:29 +00:00 · 2026-05-13 04:36:11 +00:00
parent d54ad9ca19
commit 8bac7ba0d1
2 changed files with 285 additions and 3 deletions
@@ -0,0 +1,185 @@
+/**
+ * Tests for the `seedRandomFromFrame` gate on `buildVirtualTimeShim`.
+ *
+ *   1. Backwards compatibility — `buildVirtualTimeShim({
+ *      seedRandomFromFrame: false })` returns a string byte-identical to the
+ *      legacy `VIRTUAL_TIME_SHIM`. Existing in-process callers see no
+ *      difference.
+ *
+ *   2. Distributed determinism — with `seedRandomFromFrame: true`, the shim's
+ *      `seekToTime(ms)` reseeds a Mulberry32 PRNG keyed by the virtual time
+ *      and replaces `Math.random` / `crypto.getRandomValues` with that PRNG.
+ *      `seekToTime(N)` → N `Math.random()` calls is a deterministic
+ *      sequence; reseeking to the same time restarts the sequence.
+ *
+ * The shim is executed inside `node:vm` with a synthetic `window`/`Math` so
+ * tests don't need real Chrome.
+ */
+
+import { describe, expect, it } from "bun:test";
+import { Script, createContext, type Context } from "node:vm";
+import { buildVirtualTimeShim, VIRTUAL_TIME_SHIM } from "./fileServer.js";
+
+/**
+ * Build a fresh VM context with its own globals (its own Math, its own
+ * crypto, …) and run the shim inside it. Each context's Math is independent,
+ * so we can run two shims back-to-back and have their `Math.random` overrides
+ * not clobber each other.
+ *
+ * The shim is a browser-style IIFE that touches `window.*`. We set the VM's
+ * `window` to its own globalThis so `window.setTimeout = ...` mutates the VM
+ * globals (matching browser semantics) and our test code can read
+ * `window.__HF_VIRTUAL_TIME__` afterward.
+ */
+function makeShimContext(): {
+  context: Context;
+  run: <T = unknown>(code: string) => T;
+} {
+  const context = createContext({});
+  const bootstrap = `
+    globalThis.window = globalThis;
+    globalThis.setTimeout = (cb, ms) => 0;
+    globalThis.clearTimeout = (id) => undefined;
+    globalThis.setInterval = (cb, ms) => 0;
+    globalThis.clearInterval = (id) => undefined;
+    globalThis.performance = { now: () => 0 };
+    globalThis.requestAnimationFrame = undefined;
+    globalThis.cancelAnimationFrame = undefined;
+    // Provide a synthetic crypto.getRandomValues so the shim can detect it
+    // and replace it. Default is a no-op that returns the buffer unchanged.
+    globalThis.crypto = { getRandomValues: (arr) => arr };
+  `;
+  new Script(bootstrap).runInContext(context);
+  const run = <T>(code: string): T => new Script(code).runInContext(context) as T;
+  return { context, run };
+}
+
+function runShim(shimSource: string): {
+  run: <T = unknown>(code: string) => T;
+} {
+  const ctx = makeShimContext();
+  ctx.run(shimSource);
+  return ctx;
+}
+
+describe("buildVirtualTimeShim — backwards compatibility", () => {
+  it("default (seedRandomFromFrame: false) is byte-identical to VIRTUAL_TIME_SHIM", () => {
+    // The const that existing call sites import:
+    //   renderOrchestrator.ts: preHeadScripts: [VIRTUAL_TIME_SHIM]
+    //   probeStage.ts:         preHeadScripts: [VIRTUAL_TIME_SHIM]
+    // Must continue to emit the same script.
+    const built = buildVirtualTimeShim({ seedRandomFromFrame: false });
+    expect(built).toBe(VIRTUAL_TIME_SHIM);
+  });
+
+  it("default shim does not mention any seeded-RNG identifiers", () => {
+    const shim = buildVirtualTimeShim({ seedRandomFromFrame: false });
+    expect(shim).not.toContain("mulberry32");
+    expect(shim).not.toContain("reseedRngFromTime");
+    expect(shim).not.toContain("__seededGetRandomValues");
+    expect(shim).not.toContain("Math.random = ");
+  });
+
+  it("default shim leaves Math.random pointing at the VM's native function", () => {
+    const shim = buildVirtualTimeShim({ seedRandomFromFrame: false });
+    const { run } = runShim(shim);
+    // toString() of native Math.random is `function random() { [native code] }`
+    const isNative = run<boolean>(`/\\[native code\\]/.test(Math.random.toString())`);
+    expect(isNative).toBe(true);
+  });
+});
+
+describe("buildVirtualTimeShim — seedRandomFromFrame: true", () => {
+  it("emits the seeded-RNG block", () => {
+    const shim = buildVirtualTimeShim({ seedRandomFromFrame: true });
+    expect(shim).toContain("mulberry32");
+    expect(shim).toContain("reseedRngFromTime");
+    expect(shim).toContain("__seededGetRandomValues");
+  });
+
+  it("replaces Math.random with a non-native PRNG", () => {
+    const shim = buildVirtualTimeShim({ seedRandomFromFrame: true });
+    const { run } = runShim(shim);
+    const isNative = run<boolean>(`/\\[native code\\]/.test(Math.random.toString())`);
+    expect(isNative).toBe(false);
+  });
+
+  it("produces identical Math.random sequences across two fresh VMs at the same time", () => {
+    const shim = buildVirtualTimeShim({ seedRandomFromFrame: true });
+    const drawSequence = `(() => {
+      window.__HF_VIRTUAL_TIME__.seekToTime(1234);
+      const seq = [];
+      for (let i = 0; i < 16; i++) seq.push(Math.random());
+      return seq;
+    })()`;
+    const seqA = runShim(shim).run<number[]>(drawSequence);
+    const seqB = runShim(shim).run<number[]>(drawSequence);
+    expect(seqA).toEqual(seqB);
+    // Sanity: the sequence isn't degenerate.
+    expect(new Set(seqA).size).toBeGreaterThan(8);
+    for (const v of seqA) {
+      expect(v).toBeGreaterThanOrEqual(0);
+      expect(v).toBeLessThan(1);
+    }
+  });
+
+  it("re-seeking to the same time produces the same Math.random sequence", () => {
+    // The core determinism contract for retries: same (planDir, chunkIndex) →
+    // same frame N → same seekToTime(t_N) → same Math.random outputs.
+    const shim = buildVirtualTimeShim({ seedRandomFromFrame: true });
+    const { run } = runShim(shim);
+    const observed = run<{ first: number[]; second: number[] }>(`(() => {
+      window.__HF_VIRTUAL_TIME__.seekToTime(42);
+      const first = [Math.random(), Math.random(), Math.random()];
+      window.__HF_VIRTUAL_TIME__.seekToTime(999);
+      Math.random(); Math.random();
+      window.__HF_VIRTUAL_TIME__.seekToTime(42);
+      const second = [Math.random(), Math.random(), Math.random()];
+      return { first, second };
+    })()`);
+    expect(observed.second).toEqual(observed.first);
+  });
+
+  it("different times produce different Math.random sequences", () => {
+    const shim = buildVirtualTimeShim({ seedRandomFromFrame: true });
+    const { run } = runShim(shim);
+    const observed = run<{ t0: number[]; t1: number[] }>(`(() => {
+      window.__HF_VIRTUAL_TIME__.seekToTime(0);
+      const t0 = [Math.random(), Math.random(), Math.random()];
+      window.__HF_VIRTUAL_TIME__.seekToTime(1);
+      const t1 = [Math.random(), Math.random(), Math.random()];
+      return { t0, t1 };
+    })()`);
+    expect(observed.t1).not.toEqual(observed.t0);
+  });
+
+  it("seeded crypto.getRandomValues writes deterministic bytes", () => {
+    const shim = buildVirtualTimeShim({ seedRandomFromFrame: true });
+    const drawBytes = `(() => {
+      window.__HF_VIRTUAL_TIME__.seekToTime(7);
+      const buf = new Uint8Array(64);
+      window.crypto.getRandomValues(buf);
+      return Array.from(buf);
+    })()`;
+    const a = runShim(shim).run<number[]>(drawBytes);
+    const b = runShim(shim).run<number[]>(drawBytes);
+    expect(a).toEqual(b);
+    expect(a.some((v) => v !== 0)).toBe(true);
+  });
+
+  it("seeded crypto.getRandomValues handles odd byte lengths", () => {
+    const shim = buildVirtualTimeShim({ seedRandomFromFrame: true });
+    const { run } = runShim(shim);
+    const lengths = run<number[]>(`(() => {
+      window.__HF_VIRTUAL_TIME__.seekToTime(3);
+      const out = [];
+      for (const len of [1, 2, 3, 4, 5, 7, 31, 33]) {
+        const buf = new Uint8Array(len);
+        window.crypto.getRandomValues(buf);
+        out.push(buf.byteLength);
+      }
+      return out;
+    })()`);
+    expect(lengths).toEqual([1, 2, 3, 4, 5, 7, 31, 33]);
+  });
+});
@@ -92,7 +92,96 @@ const MIME_TYPES: Record<string, string> = {
  ".otf": "font/otf",
 };

-const VIRTUAL_TIME_SHIM = String.raw`(function() {
+/**
+ * Options for {@link buildVirtualTimeShim}.
+ */
+export interface VirtualTimeShimOptions {
+  /**
+   * When `true`, the shim additionally replaces `Math.random` and
+   * `crypto.getRandomValues` with a Mulberry32-seeded PRNG keyed by the
+   * current frame's virtual time. Compositions that call `Math.random()`
+   * during render then produce byte-identical pixels across machines and
+   * across replays of the same `(planDir, chunkIndex)` pair.
+   *
+   * Default `false`: leaves `Math.random` / `crypto.getRandomValues` native,
+   * preserving the in-process renderer's non-deterministic behavior for
+   * compositions that rely on it.
+   */
+  seedRandomFromFrame: boolean;
+}
+
+/**
+ * Build the page-side virtual-time shim script.
+ *
+ * The shim freezes `Date.now`, `performance.now`, and the rAF/setTimeout
+ * pipeline so a render seek can deterministically advance the page's
+ * notion of "now". The renderer issues `__HF_VIRTUAL_TIME__.seekToTime(ms)`
+ * before every frame capture; everything timing-related on the page sees
+ * exactly `ms` until the next seek.
+ *
+ * When `options.seedRandomFromFrame` is `true`, the returned script also
+ * installs a seeded `Math.random` / `crypto.getRandomValues` keyed by the
+ * current virtual time — so compositions with stochastic visuals retry
+ * identically. When `false`, the shim emits no random-override code; the
+ * page's native `Math.random` is left alone (the in-process default).
+ */
+export function buildVirtualTimeShim(options: VirtualTimeShimOptions): string {
+  const seedRandomFromFrame = options.seedRandomFromFrame === true;
+  // The seeded-RNG block is gated at build time so the unlocked shim is
+  // byte-identical to the pre-flag form. Producer regression baselines
+  // compare on rendered pixels — but the file-server unit tests in
+  // `fileServer.test.ts` also string-match `VIRTUAL_TIME_SHIM`, and we want
+  // those matches to remain stable.
+  const seededRandomBlock = seedRandomFromFrame
+    ? String.raw`
+  // Seeded Math.random / crypto.getRandomValues, keyed by virtual time.
+  // Mulberry32 — single uint32 state, deterministic, fast.
+  var rngState = 0;
+  function mulberry32() {
+    rngState |= 0; rngState = (rngState + 0x6D2B79F5) | 0;
+    var t = rngState;
+    t = Math.imul(t ^ (t >>> 15), t | 1);
+    t ^= t + Math.imul(t ^ (t >>> 7), t | 61);
+    return ((t ^ (t >>> 14)) >>> 0) / 4294967296;
+  }
+  function reseedRngFromTime(ms) {
+    var ms32 = Math.max(0, Math.floor(Number(ms) || 0)) | 0;
+    // Knuth's multiplicative hash + golden-ratio offset — gives a well-
+    // distributed seed even for frame 0 (otherwise rngState=0 degenerates
+    // the PRNG's first few outputs).
+    rngState = (Math.imul(ms32, -1640531527) + 0x9E3779B9) | 0;
+  }
+  reseedRngFromTime(0);
+  try {
+    Math.random = function() { return mulberry32(); };
+  } catch (e) {}
+  if (window.crypto && typeof window.crypto.getRandomValues === "function") {
+    try {
+      var __seededGetRandomValues = function(arr) {
+        if (!arr || typeof arr.byteLength !== "number" || !arr.buffer) return arr;
+        var byteLen = arr.byteLength;
+        if (byteLen <= 0) return arr;
+        var view = new DataView(arr.buffer, arr.byteOffset, byteLen);
+        var i = 0;
+        for (; i + 4 <= byteLen; i += 4) {
+          var word = ((mulberry32() * 4294967296) >>> 0);
+          view.setUint32(i, word, true);
+        }
+        for (; i < byteLen; i++) {
+          view.setUint8(i, (mulberry32() * 256) | 0);
+        }
+        return arr;
+      };
+      window.crypto.getRandomValues = __seededGetRandomValues;
+    } catch (e) {}
+  }
+`
+    : "";
+  // The seekToTime hook reseeds when seeding is on; under seedRandomFromFrame=false
+  // we emit no extra call so the function body is byte-identical to the
+  // unseeded shim.
+  const seekToTimeReseedCall = seedRandomFromFrame ? "reseedRngFromTime(safeTimeMs);\n      " : "";
+  return String.raw`(function() {
  if (window.__HF_VIRTUAL_TIME__) return;

  var virtualNowMs = 0;
@@ -109,7 +198,7 @@ const VIRTUAL_TIME_SHIM = String.raw`(function() {
  var originalCancelAnimationFrame = window.cancelAnimationFrame
    ? window.cancelAnimationFrame.bind(window)
    : null;
-
+${seededRandomBlock}
  function flushAnimationFrame() {
    if (!rafQueue.length) return;
    var current = rafQueue.slice();
@@ -180,7 +269,7 @@ const VIRTUAL_TIME_SHIM = String.raw`(function() {
    seekToTime: function(nextTimeMs) {
      var safeTimeMs = Math.max(0, Number(nextTimeMs) || 0);
      virtualNowMs = safeTimeMs;
-      flushAnimationFrame();
+      ${seekToTimeReseedCall}flushAnimationFrame();
      return virtualNowMs;
    },
    getTime: function() {
@@ -188,6 +277,14 @@ const VIRTUAL_TIME_SHIM = String.raw`(function() {
    },
  };
 })();`;
+}
+
+/**
+ * Default in-process virtual-time shim — `seedRandomFromFrame: false`.
+ * Existing call sites (`renderOrchestrator`, `probeStage`) import this
+ * constant. Distributed callers build their own with seeding enabled.
+ */
+const VIRTUAL_TIME_SHIM = buildVirtualTimeShim({ seedRandomFromFrame: false });

 /**
 * Render mode extension -- adds renderSeek() for frame-accurate seeking