Files
PaperClipAI/scripts/native-ingestion-performance-gates.test.mjs
T

203 lines
17 KiB
JavaScript

import assert from "node:assert/strict";
import { spawnSync } from "node:child_process";
import { createHash } from "node:crypto";
import { mkdtemp, readFile, rm, writeFile } from "node:fs/promises";
import os from "node:os";
import path from "node:path";
import test from "node:test";
import { evaluateServiceRepeatability, noiseExceedingPairedWins } from "./native-ingestion-performance-gates.mjs";
test("four pairs must individually improve beyond noise", () => {
const threeBeyondNoise = noiseExceedingPairedWins([0.30, 0.30, 0.30, 0.01, -0.01], 0.10);
assert.equal(threeBeyondNoise.noiseExceedingPairedWins, 3);
assert.equal(threeBeyondNoise.pairedWinRequirementPassed, false);
assert.equal(noiseExceedingPairedWins([0.30, 0.25, 0.20, 0.15, 0.01], 0.10).pairedWinRequirementPassed, true);
assert.equal(noiseExceedingPairedWins([0.30, 0.30, 0.30, 0.10, -0.01], 0.10).pairedWinRequirementPassed, false);
});
function repeatabilityReport() {
return { phase: "aa", incomplete: false, cleanupComplete: true,
workload: { quick: false, pairs: 5, events: 32, groups: 16, histories: [1000, 10000, 100000],
scenarios: ["ordered", "replay", "late-gap", "warm-suffix"] },
results: [1000, 10000, 100000].flatMap(history => ["ordered", "replay", "late-gap", "warm-suffix"].map(scenario => ({ history, scenario,
baseline: { samples: Array.from({ length: 5 }, () => ({ elapsedMs: 100 })) },
candidate: { samples: Array.from({ length: 5 }, () => ({ elapsedMs: 101 })) },
}))) };
}
test("service A/A enforces both preregistered limits on every timing case", () => {
assert.equal(evaluateServiceRepeatability(repeatabilityReport()).passed, true);
const variance = repeatabilityReport(); variance.results[11].candidate.samples[0].elapsedMs = 121;
assert.equal(evaluateServiceRepeatability(variance).passed, false);
const drift = repeatabilityReport(); drift.results[0].candidate.samples.forEach(value => { value.elapsedMs = 111; });
assert.equal(evaluateServiceRepeatability(drift).passed, false);
const boundary = repeatabilityReport(); boundary.results[0].candidate.samples.forEach(value => { value.elapsedMs = 110; });
boundary.results[0].candidate.samples[0].elapsedMs = 120;
assert.equal(evaluateServiceRepeatability(boundary).passed, true, "The stated inclusive 20%/10% boundaries qualify");
const late = repeatabilityReport(); late.results[2].candidate.samples[0].elapsedMs = 1000;
assert.equal(evaluateServiceRepeatability(late).passed, true, "Late-gap time stays structural-only");
});
test("smoke, partial, missing matrix, insufficient pairs and unfinished cleanup cannot qualify A/A", () => {
for (const change of [report => { report.workload.quick = true; }, report => { report.incomplete = true; },
report => { report.cleanupComplete = false; }, report => { report.workload.pairs = 4; },
report => { report.results.pop(); }, report => { report.phase = "compare"; }]) {
const report = repeatabilityReport(); change(report);
assert.equal(evaluateServiceRepeatability(report).passed, false);
}
});
test("owner report-validation CLI enforces repeatability without opening a database", async () => {
const directory = await mkdtemp(path.join(os.tmpdir(), "native-ingestion-owner-gates-"));
const root = path.resolve(import.meta.dirname, "..");
const report = repeatabilityReport();
report.candidateSourceHashes = Object.fromEntries(await Promise.all([
"scripts/benchmark-native-event-ingestion.ts", "scripts/native-ingestion-performance-gates.mjs",
].map(async file => [file, createHash("sha256").update(await readFile(path.join(root, file))).digest("hex")])));
const file = path.join(directory, "report.json");
const run = () => spawnSync(process.execPath, ["--import", "./server/node_modules/tsx/dist/loader.mjs",
"scripts/benchmark-native-event-ingestion.ts", `--validate-report=${file}`, "--require-repeatability"], { cwd: root, encoding: "utf8" });
try {
await writeFile(file, JSON.stringify(report));
const valid = run(); assert.equal(valid.status, 0, valid.stderr);
report.results[0].candidate.samples[0].elapsedMs = 150;
await writeFile(file, JSON.stringify(report));
const invalid = run(); assert.notEqual(invalid.status, 0); assert.match(invalid.stderr, /repeatability gate failed/);
} finally { await rm(directory, { recursive: true, force: true }); }
});
const workloads = [
{ name: "paced", runs: 1, events: 128, intervalMs: 25, prelude: 0 },
{ name: "burst16", runs: 1, events: 16, intervalMs: 0, prelude: 0 },
{ name: "burst128", runs: 10, events: 128, intervalMs: 0, prelude: 0 },
{ name: "burst512", runs: 100, events: 512, intervalMs: 0, prelude: 0 },
{ name: "existing-history", runs: 10, events: 128, intervalMs: 0, prelude: 1000 },
{ name: "operator-stop", runs: 1, events: 512, intervalMs: 100, prelude: 0, stop: true },
];
function browserReport(revision, latency) {
const companyId = "00000000-0000-4000-8000-000000000001", environmentId = "00000000-0000-4000-8000-000000000002";
const id = (kind, caseIndex, index) => `00000000-0000-4000-8000-${(kind * 100000 + caseIndex * 1000 + index).toString(16).padStart(12, "0")}`;
return { protocolVersion: "v3-supported-remote-original-clock-stream", transport: "sandbox-remote", companyId,
resourceBudgets: { providerReadinessMs: 300000, fullInvocationMs: 900000, streamingMs: 120000,
terminalDrainageMs: 180000, stopAcknowledgementMs: 30000, remoteLeaseCleanupMs: 30000 },
remote: { transport: "sandbox-remote", revision, buildRevision: revision, environmentId,
runnerPublicUrl: "wss://127.0.0.1:32100",
selection: "Normal plugin install + sandbox environment + agent defaultEnvironmentId; no runtime resolver override",
plugin: { pluginKey: "paperclip.fake-sandbox-provider" }, tls: { peerVerification: true, caSha256: "e".repeat(64), serverCertificateSha256: "f".repeat(64) },
runner: { sha256: "d".repeat(64), metadata: { prpTransportModes: ["dial_wss"] } },
provider: { nodeSha256: "b".repeat(64), wrapperSha256: "c".repeat(64), fixtureSha256: "a".repeat(64) } },
fixtureSha256: "a".repeat(64), admissionProtocol: "sequential-provider-and-clock-readiness-before-common-epoch", setupReadinessPollingMs: 50, complete: true, cleanup: { complete: true, errors: [] }, qualificationOnly: false, diagnosticOnly: false, performanceAcceptanceEligible: true, browserRecording: { trace: "off", video: "off", screencast: false },
sourceRevision: revision, compiledBuildRevision: revision, serverArtifact: "compiled", serverEntrySha256: `${revision}-compiled`,
measurementSha256: "synthetic-fixed-fixture", compiledHotPathHashes: { append: "append", port: "port", runtime: "runtime" },
cases: workloads.map((workload, caseIndex) => ({ ...workload, stop: Boolean(workload.stop), persistedAfterReload: true,
epochMs: 1000, preludeProof: workload.prelude ? { individualDeltas: workload.prelude, uniqueSourceIdentities: workload.prelude, verifiedAtMs: 999 } : undefined,
settledAtMs: 1000 + latency, stopRequestedAtMs: workload.stop ? 1000 : undefined,
stopAcknowledgedAtMs: workload.stop ? 1000 + latency : undefined, remoteLeasesCheckedAtMs: 1001 + latency,
runIds: Array.from({ length: workload.runs }, (_, index) => id(1, caseIndex, index)),
controllerAuthority: { checkedBeforeMeasuredEpoch: true, checkedAtMs: 998, connectionClosedAtMs: 999,
observedReadOnly: "on", observedIsolation: "repeatable read", rows: Array.from({ length: workload.runs }, (_, index) => ({
run_id: id(1, caseIndex, index), issue_id: id(2, caseIndex, index), company_id: companyId,
lease_id: id(3, caseIndex, index), environment_id: environmentId, runtime_mode: "native", provider: "fake-plugin",
provider_lease_id: `fake-plugin://${id(1, caseIndex, index)}/fixture`, checkpoint_part: "header", checkpoint_header_bytes: 256,
})) },
remoteLeases: Array.from({ length: workload.runs }, (_, index) => ({ id: id(3, caseIndex, index), runId: id(1, caseIndex, index),
companyId, environmentId, provider: "fake-plugin", providerLeaseId: `fake-plugin://${id(1, caseIndex, index)}/fixture`,
status: "released", cleanupStatus: "success", releasedAt: "2026-11-03T00:00:00.000Z" })),
visible: Array(workload.stop ? 3 : workload.events).fill({}), matchedVisibleClocks: workload.stop ? 3 : workload.events,
firstVisibleAfterEmissionMs: latency, emissionToVisible: { medianMs: latency, p95Ms: latency },
interChunkGaps: { paintCount: workload.events, count: workload.events - 1, medianMs: latency, maxMs: latency },
intendedToVisible: { p95Ms: latency }, drainThroughTerminalMs: latency,
stopThroughTerminalMs: latency, stopThroughAcknowledgementMs: latency,
durable: Array.from({ length: workload.runs }, (_, index) => ({ runId: id(1, caseIndex, index), token: String(index), ordered: true,
uniqueSourceIdentities: workload.stop ? 3 : workload.events, individualDeltas: workload.stop ? 3 : workload.events, matchedSourceClocks: workload.stop ? 3 : workload.events })),
})) };
}
test("browser CLI rejects the three-pairs-above-noise counterexample and ineligible evidence", async () => {
const directory = await mkdtemp(path.join(os.tmpdir(), "native-ingestion-gates-"));
const script = new URL("./compare-native-ingestion-browser.ts", import.meta.url);
const save = async (name, value) => { const file = path.join(directory, name); await writeFile(file, JSON.stringify(value)); return file; };
async function manifest(prefix, improvements, sameRevision = false, mutate) {
const entries = [];
for (let index = 0; index < 5; index++) {
const baseline = browserReport("baseline", 1000), candidate = browserReport(sameRevision ? "baseline" : "candidate", 1000 * (1 - improvements[index]));
if (mutate && index === 0) mutate(candidate);
entries.push({ baselineFile: await save(`${prefix}-${index}-baseline.json`, baseline),
candidateFile: await save(`${prefix}-${index}-candidate.json`, candidate) });
}
return save(`${prefix}-pairs.json`, entries);
}
const run = args => spawnSync(process.execPath, [script.pathname, ...args], { encoding: "utf8" });
try {
const noise = path.join(directory, "noise.json");
const aa = run(["--phase=aa", `--pairs=${await manifest("aa", Array(5).fill(0.10), true)}`, `--output=${noise}`]);
assert.equal(aa.status, 0, aa.stderr);
const badOutput = path.join(directory, "three-wins.json");
const bad = run([`--pairs=${await manifest("three-wins", [0.30, 0.30, 0.30, 0.01, -0.01])}`, `--noise-report=${noise}`, `--output=${badOutput}`, "--require-acceptance"]);
assert.notEqual(bad.status, 0);
const result = JSON.parse(await readFile(badOutput, "utf8"));
assert.equal(result.acceptancePassed, false);
assert.ok(result.gates.filter(gate => gate.improvementRequired).every(gate => gate.noiseExceedingPairedWins === 3));
const good = run([`--pairs=${await manifest("four-wins", [0.30, 0.25, 0.20, 0.15, 0.01])}`, `--noise-report=${noise}`, "--require-acceptance"]);
assert.equal(good.status, 0, good.stderr);
for (const [name, change, message] of [
["smoke", report => { report.qualificationOnly = true; }, /Smoke cannot qualify/],
["source", report => { report.serverArtifact = "source-qualification"; }, /compiled server/],
["profile", report => { report.diagnosticOnly = true; }, /Diagnostic profiles/],
["profile-ineligible", report => { report.performanceAcceptanceEligible = false; }, /Profile-only evidence/],
["missing-retained-prelude", report => { report.cases.find(row => row.prelude).preludeProof.individualDeltas--; }, /stored prelude deltas/],
["legacy-protocol", report => { delete report.protocolVersion; }, /v3 remote clock protocol/],
["old-local-protocol", report => { report.protocolVersion = "v2-compact-markers-original-clock-stream"; report.transport = "local"; }, /v3 remote clock protocol/],
["wrong-transport", report => { report.transport = "local"; }, /sandbox remote transport/],
["unregistered-resource-budget", report => { report.resourceBudgets.providerReadinessMs = 600000; }, /registered remote setup\/resource budgets/],
["missing-normal-remote", report => { delete report.remote; }, /normal remote runtime provenance/],
["missing-plugin", report => { delete report.remote.plugin; }, /plugin provenance/],
["resolver-override", report => { report.remote.selection = "forced external controller flag"; }, /normal API remote environment selection/],
["tls-unverified", report => { report.remote.tls.peerVerification = false; }, /TLS peer verification/],
["plaintext-remote", report => { report.remote.runnerPublicUrl = "ws://127.0.0.1:32100"; }, /endpoint must use verified WSS/],
["foreign-remote", report => { report.remote.runnerPublicUrl = "wss://example.com:32100"; }, /owned loopback/],
["different-provider-artifact", report => { report.remote.provider.nodeSha256 = "9".repeat(64); }, /artifact changed/],
["fixed-workload-changed", report => { report.cases[3].runs = 10; }, /Fixed workload changed/],
["missing-controller-authority", report => { delete report.cases[3].controllerAuthority; }, /controller authority proof/],
["authority-writable", report => { report.cases[3].controllerAuthority.observedReadOnly = "off"; }, /read-only transaction/],
["authority-wrong-isolation", report => { report.cases[3].controllerAuthority.observedIsolation = "read committed"; }, /repeatable-read isolation/],
["authority-late", report => { report.cases[3].controllerAuthority.checkedAtMs = 1000; }, /before the measured epoch/],
["authority-fractional-closed-time", report => { report.cases[3].controllerAuthority.connectionClosedAtMs = 999.5; }, /before the measured epoch/],
["authority-missing-run", report => { report.cases[3].controllerAuthority.rows.pop(); }, /Every remote run/],
["authority-wrong-run", report => { report.cases[3].controllerAuthority.rows[99].run_id = report.cases[3].controllerAuthority.rows[0].run_id; }, /run set mismatch/],
["authority-wrong-company", report => { report.cases[3].controllerAuthority.rows[99].company_id = "other"; }, /company mismatch/],
["authority-wrong-environment", report => { report.cases[3].controllerAuthority.rows[99].environment_id = "other"; }, /environment mismatch/],
["authority-wrong-provider", report => { report.cases[3].controllerAuthority.rows[99].provider = "fake"; }, /fake-plugin/],
["authority-missing-checkpoint", report => { report.cases[3].controllerAuthority.rows[99].checkpoint_part = null; }, /checkpoint header/],
["missing-lease-cleanup", report => { delete report.cases[3].remoteLeases; }, /lease cleanup proof/],
["lease-pending", report => { report.cases[3].remoteLeases[99].status = "pending_cleanup"; }, /cleanup remains unsettled/],
["lease-wrong-binding", report => { report.cases[3].remoteLeases[99].providerLeaseId = "other"; }, /fake-plugin/],
["lease-check-before-timing", report => { report.cases[3].remoteLeasesCheckedAtMs = report.cases[3].settledAtMs - 1; }, /after recorded timing boundaries/],
["stop-timing-restamped", report => { report.cases[5].stopAcknowledgedAtMs++; }, /timing boundary changed/],
["unqualified-admission", report => { delete report.admissionProtocol; }, /admission protocol/],
["unqualified-readiness-observation", report => { delete report.setupReadinessPollingMs; }, /readiness observation interval/],
["unfinished-cleanup", report => { report.cleanup.complete = false; }, /completed fixture cleanup/],
["cleanup-error", report => { report.cleanup.errors.push({ step: "restore", message: "timeout" }); }, /cleanup errors/],
["missing-paced-gaps", report => { delete report.cases[0].interChunkGaps; }, /inter-chunk paint gaps/],
["unmatched-visible-clock", report => { report.cases[0].matchedVisibleClocks--; }, /original provider clock/],
["unmatched-durable-clock", report => { report.cases[0].durable[0].matchedSourceClocks--; }, /original provider clock/],
["browser-trace", report => { report.browserRecording.trace = "on"; }, /trace capture off/],
["browser-video", report => { report.browserRecording.video = "on"; }, /video capture off/],
["browser-screencast", report => { report.browserRecording.screencast = true; }, /screencast capture off/],
]) {
const rejected = run([`--pairs=${await manifest(name, Array(5).fill(0.30), false, change)}`, `--noise-report=${noise}`, "--require-acceptance"]);
assert.notEqual(rejected.status, 0); assert.match(rejected.stderr, message);
}
for (const [name, change, message] of [
["old-noise-protocol", report => { report.protocolVersion = "v2-compact-markers-original-clock-stream"; }, /different remote protocol/],
["local-noise-transport", report => { report.transport = "local"; }, /different transport/],
["changed-noise-budget", report => { report.resourceBudgets.providerReadinessMs++; }, /budgets changed after A\/A/],
]) {
const changed = JSON.parse(await readFile(noise, "utf8")); change(changed);
const file = await save(name + ".json", changed);
const rejected = run([`--pairs=${await manifest(name, Array(5).fill(0.30))}`, `--noise-report=${file}`, "--require-acceptance"]);
assert.notEqual(rejected.status, 0); assert.match(rejected.stderr, message);
}
} finally { await rm(directory, { recursive: true, force: true }); }
});