[campaign #239] STEP 4-B — STDLIB FILL III: randomness + time bindings

[hyperpolymath]

Part of #239, follow-up to #242 (STEP 3) and pre-req for #243 (STEP 4 per-repo ports).

Why this is needed

The STEP 4 triage on #243 found that bofig's property tests (tests/property/graph_properties_test.ts, 377L) generate test inputs with:

function randomId(prefix: string = "node"): string {
  return `${prefix}_${Math.floor(Math.random() * 1000000)}`;
}

Math.random() is not bound in AffineScript's Deno-ESM surface today. Same pattern likely recurs in any future property-test port that uses generator-driven fuzzing rather than fixed-seed inputs.

Required surface

Minimum viable randomness:

Extern	Lowers to	Notes
math_random() -> Float	Math.random()	[0, 1). JS PRNG, non-crypto.
random_u32() -> Int	(Math.random() * 0x100000000) >>> 0	Convenience: uniform u32.
random_in_range(lo: Int, hi: Int) -> Int	Math.floor(Math.random() * (hi - lo)) + lo	Uniform [lo, hi).

If property tests need crypto-grade randomness later (e.g. cryptographic key generation), file a separate crypto_random_bytes binding sub-issue under #239 — it's a different host call (crypto.getRandomValues) and a different threat model.

Time bindings the same triage flagged as occasional needs:

Extern	Lowers to	Notes
performance_now() -> Float	performance.now()	High-res timer (sub-ms). Existing dateNow() -> Int millis is fine for coarse timings; bench tests want sub-ms resolution.

Out of scope

• Seeded RNG (@std/random) — defer until a property-test port actually needs determinism.
• sleep / setTimeout — separate effect surface; not blocking STEP 4 tests.

Acceptance

• Externs ship in stdlib/Deno.affine with docstrings.
• tests/codegen-deno/random_smoke.{affine,deno.js,harness.mjs} exercises distribution sanity (random_u32 returns ≥ 1000 distinct values in 10000 draws; random_in_range(0, 100) stays in [0, 100) over 1000 draws; performance_now returns a monotone non-decreasing number across two calls).
• dune runtest green; tools/run_codegen_deno_tests.sh green.

Refs

• Umbrella: [campaign] TypeScript → AffineScript estate migration (UMBRELLA) #239
• Predecessor: [campaign #239] STEP 3 — STDLIB FILL: discharge gaps surfaced by step 2 #242 (STEP 3) + affinescript#504
• Sibling: [campaign #239] STEP 4-A — STDLIB FILL II: raw binary I/O (Bytes constructor + LE getters/setters) #326 (STEP 4-A binary I/O)
• Direct unblock: [campaign #239] STEP 4 — TAIL BATCH 2: mid-tier pure-logic repos (4-9 files) #243 (STEP 4) — specifically bofig property/bench files

[hyperpolymath]

PR filed: hyperpolymath/affinescript#509 (auto-merge SQUASH armed). 4 externs (math_random / random_u32 / random_in_range / performance_now) + lowerings + random_smoke test fixture validates distribution sanity (1000 unit draws in [0,1); 10000 u32 draws all in u32 with ≥1000 distinct; 1000 in_range(0,100) draws all in [0,100); monotone perf.now). 353/353 tests green. Closes this issue on merge.