josh/ytplayer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'agy/phase0-perf'

Browse Source
This commit is contained in:
Jonathan Sykes
2026-10-07 16:36:56 +08:00
parent 233bf94e4c 4bc15a55e3
commit 7bc5b4f8b5
7 changed files with 3225 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
.gitignore vendored
View File

@@ -40,3 +40,7 @@ legacy/releases/
__pycache__/ __pycache__/
plans/codex-shots*/ plans/codex-shots*/
plans/codex-report*.md plans/codex-report*.md
# performance harness artifacts and temp files
perf/fixtures/
perf/.tmp/

132
perf/README.md Normal file
View File

@@ -0,0 +1,132 @@
# Performance Baseline Harness (`perf/`)
Automated performance baseline and regression harness for the **ytplayer PWA**.
The harness drives the real Bun backend server (`server/server.js`) serving static Brotli-compressed assets, combined with Playwright browser automation on Chromium and WebKit.
---
## 1. Architecture
### Backend Server (`server/server.js`)
- Runs as an isolated child process using **Bun** on an ephemeral port.
- Configured with dedicated scratch directories under `perf/.tmp/` (`DATA_DIR`, `DB_PATH`, `MEDIA_DIR`) so no production or developer data is ever touched.
- Employs the real static asset delivery stack (`Brotli` encoding, `ETag` generation, `Cache-Control` immutable caching for version-stamped assets, and `computeBuildTag()`).
### Browser Automation & Network Throttling
- **Chromium**:
- Network emulation via Chrome DevTools Protocol (`Network.emulateNetworkConditions`).
- CPU throttling at **4× slowdown** (`Emulation.setCPUThrottlingRate`).
- Unrestricted audio/video playback via `--autoplay-policy=no-user-gesture-required`.
- **WebKit**:
- WebKit on Linux lacks CDP network emulation capabilities.
- Throttling and stalling are handled by a dedicated local proxy (`perf/proxy.mjs`) sitting between WebKit and the Bun server.
- The proxy shapes traffic by pacing chunk delivery through backpressured streams and injecting round-trip latency and stalls.
- **Network Profiles**:
- **`lte`**: 1.6 Mbps (200 KB/s) download, 750 kbps upload, 150 ms round-trip latency.
- **`lossy`**: 1.0 Mbps (125 KB/s) download, 250 ms latency, with intermittent connection stalls (every 7th request delayed by 1,500 ms).
### Shell Fixture Extractor (`perf/make-shell-fixture.sh`)
- Archives the frontend from git commits (`git archive <commit> frontend`) into `perf/fixtures/shell-<tag>/`.
- Computes the build tag using the exact sha256 walk algorithm from `server/server.js computeBuildTag()`.
- Validates the computed tag against live production (`https://worship.hesed.sbs/api/version`).
---
## 2. Test Scenarios
1. **`cold` (Cold First Load)**:
- Fresh browser context with empty cache, no service worker, and clean storage.
- Measures initial page fetch, FCP, LCP, DOMContentLoaded, boot milestone (`#cards` render), media playback readiness, service worker precaching duration, and origin storage quota.
2. **`warm` (Warm Reload under SW)**:
- Re-navigates the page with active Service Worker and primed HTTP cache.
- Measures cached asset retrieval efficiency and second-load boot performance.
3. **`offline` (Offline Reload)**:
- Sets offline state and reloads the application.
- Verifies the app boots entirely from `CacheStorage` precached shell assets without network connectivity.
4. **`update-js` (Update N → N+1 with 1-line JS comment)**:
- Boots client against Tree N (base frontend), establishing an active Service Worker.
- Swaps server to Tree N+1 (one comment line appended to `app.js`, altering `BUILD_TAG`).
- Triggers `registration.update()` and measures total wire bytes and time until the new Service Worker reaches `waiting` state (`installed`).
5. **`update-css` (Update N → N+1 with 1-line CSS rule)**:
- Identical to `update-js`, but Tree N+1 has one CSS rule appended to `theme-glass.css`.
- Measures the full shell re-download penalty caused by any stylesheet modification.
---
## 3. Metrics Collected
Each scenario records the following metrics across **5 independent runs** (reporting min, max, and median):
- **Request Count**: Total HTTP requests made across the wire.
- **Wire Bytes**: Total compressed bytes transferred over the network.
- **Wire Bytes by Type**: Categorized byte breakdown (`html`, `js`, `css`, `fonts`, `images`, `media`, `api`, `other`).
- **FCP (First Contentful Paint)**: Time from navigation start to first rendered content (ms).
- **LCP (Largest Contentful Paint)**: Time to largest contentful paint via `PerformanceObserver` (ms).
- **DOMContentLoaded**: Time to DOMContentLoaded event end (ms).
- **Boot Done**: Time from navigation start until the first `#cards` view rendering is completed (ms).
- **Long Tasks Total**: Cumulative execution time of main thread tasks exceeding 50 ms (ms, Chromium).
- **SW Install Duration**: Time from Service Worker registration until precaching completes (ms).
- **SW Install Wire Bytes**: Network bytes transmitted during Service Worker installation.
- **Storage Estimate**: Quota and usage reported by `navigator.storage.estimate()` (bytes).
- **Media Ready**: Elapsed time for deterministic media fixture (`benchmedia01`) to reach `readyState >= 3` / playing state. Notes if autoplay policy blocked playback.
---
## 4. How to Run
### Prerequisites
Symlink dependencies and ensure server dependencies are installed:
```bash
# Worktree root
ln -s ~/development/personal/ytplayer/node_modules node_modules
cd server && bun install && cd ..
```
### Basic Commands
Run the complete baseline suite (5 runs, all browsers, LTE profile):
```bash
node perf/baseline.mjs
```
Run a specific browser or profile:
```bash
# Chromium on LTE profile
node perf/baseline.mjs --browser chromium --profile lte
# WebKit on LTE profile
node perf/baseline.mjs --browser webkit --profile lte
# Both browsers on lossy profile
node perf/baseline.mjs --browser all --profile lossy
```
Run a single scenario:
```bash
# Cold load only
node perf/baseline.mjs --scenario cold
# Update scenario with 1-line JS change
node perf/baseline.mjs --scenario update-js
# Update scenario with 1-line CSS change
node perf/baseline.mjs --scenario update-css
```
Configure run count:
```bash
# Quick sanity check (1 iteration)
node perf/baseline.mjs --runs 1
# High precision benchmark (7 iterations)
node perf/baseline.mjs --runs 7
```
---
## 5. Phase 5 Comparison Mode (`--compare`)
To verify performance gains and detect regressions in later phases:
```bash
node perf/baseline.mjs --compare perf/results/baseline-2026-10-07.json
```
Prints an ASCII/Markdown delta comparison table highlighting percentage changes across wire bytes, boot duration, and update overhead.

948
perf/baseline.mjs Executable file
View File

@@ -0,0 +1,948 @@
#!/usr/bin/env node
/**
* perf/baseline.mjs
*
* Performance baseline harness for ytplayer PWA.
* Drives the real Bun server and runs Playwright (Chromium with CDP throttling + CPU 4x;
* WebKit with a local throttling & stalling proxy).
*
* Scenarios:
* - cold: fresh context, network download, SW install, boot, media playback check
* - warm: re-navigate with active SW and primed HTTP cache
* - offline: context.setOffline(true), reload under SW cache
* - update-js: update N->N+1 (1-line comment appended to app.js) -> wire bytes + time until new SW ready
* - update-css: update N->N+1 (1-line rule appended to theme-glass.css) -> wire bytes + time until new SW ready
*
* Usage:
* node perf/baseline.mjs [options]
* Options:
* --runs <number> Runs per scenario for median calculation (default: 5)
* --browser <chr|wk|all> Browser(s) to test (default: all)
* --profile <lte|lossy|all> Network profile (default: lte)
* --scenario <name|all> Specific scenario or all (default: all)
* --out <path> Output JSON path
* --compare <json-path> Compare results against a previous baseline JSON
*/
import { spawn, spawnSync } from 'node:child_process';
import { existsSync, mkdirSync, rmSync, cpSync, appendFileSync, statSync, writeFileSync, readFileSync, symlinkSync, unlinkSync } from 'node:fs';
import path from 'node:path';
import { fileURLToPath } from 'node:url';
import { createThrottleProxy } from './proxy.mjs';
const __filename = fileURLToPath(import.meta.url);
const __dirname = path.dirname(__filename);
const REPO_ROOT = path.resolve(__dirname, '..');
// Dynamically resolve playwright from node_modules
let pw;
try {
pw = await import('playwright');
} catch {
pw = await import(path.join(REPO_ROOT, 'node_modules', 'playwright', 'index.mjs'));
}
import { createRequire } from 'node:module';
const req = createRequire(import.meta.url);
const { createClient } = req(path.join(REPO_ROOT, 'server', 'node_modules', '@libsql/client'));
const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms));
function parseArgs() {
const args = process.argv.slice(2);
const options = {
runs: 5,
browser: 'all', // 'chromium', 'webkit', or 'all'
profile: 'lte', // 'lte', 'lossy', or 'all'
scenario: 'all', // 'cold', 'warm', 'offline', 'update-js', 'update-css', or 'all'
out: null,
compare: null,
};
for (let i = 0; i < args.length; i++) {
const a = args[i];
if (a === '--runs' && i + 1 < args.length) options.runs = parseInt(args[++i], 10);
else if (a === '--browser' && i + 1 < args.length) options.browser = args[++i];
else if (a === '--profile' && i + 1 < args.length) options.profile = args[++i];
else if (a === '--scenario' && i + 1 < args.length) options.scenario = args[++i];
else if (a === '--out' && i + 1 < args.length) options.out = args[++i];
else if (a === '--compare' && i + 1 < args.length) options.compare = args[++i];
}
return options;
}
function median(values) {
const nums = values.filter((v) => typeof v === 'number' && !Number.isNaN(v)).sort((a, b) => a - b);
if (!nums.length) return null;
const mid = Math.floor(nums.length / 2);
return nums.length % 2 !== 0 ? nums[mid] : Math.round(((nums[mid - 1] + nums[mid]) / 2) * 10) / 10;
}
function summarizeList(values) {
const valid = values.filter((v) => v !== null && v !== undefined && !Number.isNaN(v));
if (!valid.length) return { median: null, min: null, max: null, values };
return {
median: median(valid),
min: Math.min(...valid),
max: Math.max(...valid),
values,
};
}
/**
* Setup deterministic local media fixture
*/
function prepareMediaFiles(mediaDir) {
mkdirSync(mediaDir, { recursive: true });
const mp4 = path.join(mediaDir, 'benchmedia01.1.mp4');
const m4a = path.join(mediaDir, 'benchmedia01.1.m4a');
if (!existsSync(mp4)) {
spawnSync('ffmpeg', [
'-y', '-f', 'lavfi', '-i', 'testsrc=size=640x360:rate=30:duration=5',
'-f', 'lavfi', '-i', 'sine=frequency=440:duration=5',
'-c:v', 'libx264', '-preset', 'ultrafast', '-pix_fmt', 'yuv420p',
'-c:a', 'aac', '-movflags', '+faststart', '-shortest', mp4,
]);
}
if (!existsSync(m4a)) {
spawnSync('ffmpeg', [
'-y', '-f', 'lavfi', '-i', 'sine=frequency=440:duration=5',
'-c:a', 'aac', '-shortest', m4a,
]);
}
return { mp4, m4a };
}
async function seedDatabase(dbPath, mediaFiles) {
const db = createClient({ url: 'file:' + dbPath });
const mp4Size = statSync(mediaFiles.mp4).size;
const m4aSize = statSync(mediaFiles.m4a).size;
await db.execute({
sql: `INSERT OR REPLACE INTO media_cache (
video_id, status, gen, size, height, vcodec, acodec, duration, meta, last_access
) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?)`,
args: [
'benchmedia01',
'ready',
1,
mp4Size + m4aSize,
360,
'h264',
'aac',
5.0,
JSON.stringify({ title: 'Baseline Deterministic Clip', channel: 'Perf Benchmark' }),
Date.now(),
],
});
}
/**
* Bun Server Manager
*/
class ServerManager {
constructor(scratchDir, publicDir) {
this.scratchDir = scratchDir;
this.publicDir = publicDir;
this.dataDir = path.join(scratchDir, 'data');
this.mediaDir = path.join(this.dataDir, 'media');
this.dbPath = path.join(this.dataDir, 'ytplayer.db');
this.serverProcess = null;
this.port = 0;
}
async start() {
mkdirSync(this.dataDir, { recursive: true });
const fixtureDir = path.join(REPO_ROOT, 'perf', 'fixtures', 'media');
const mediaFiles = prepareMediaFiles(fixtureDir);
// Create a dedicated server working dir with public symlink
const srvCwd = path.join(this.scratchDir, 'srv');
mkdirSync(srvCwd, { recursive: true });
const publicLink = path.join(srvCwd, 'public');
if (existsSync(publicLink)) unlinkSync(publicLink);
symlinkSync(this.publicDir, publicLink);
// Pick random available ephemeral port once and reuse across restarts
this.port = this.port || (4300 + Math.floor(Math.random() * 2000));
const env = {
...process.env,
PORT: String(this.port),
DATA_DIR: this.dataDir,
MEDIA_DIR: this.mediaDir,
DB_PATH: this.dbPath,
};
const serverScript = path.join(REPO_ROOT, 'server', 'server.js');
this.serverProcess = spawn('bun', [serverScript], {
cwd: srvCwd,
env,
stdio: 'ignore',
});
// Wait for server readiness
let ready = false;
for (let i = 0; i < 50; i++) {
try {
const res = await fetch(`http://127.0.0.1:${this.port}/api/version`);
if (res.ok) {
ready = true;
break;
}
} catch {}
await sleep(100);
}
if (!ready) {
throw new Error(`Server failed to start on port ${this.port}`);
}
// Copy media files to mediaDir and seed media table AFTER server startup sweep
mkdirSync(this.mediaDir, { recursive: true });
const targetMp4 = path.join(this.mediaDir, 'benchmedia01.1.mp4');
const targetM4a = path.join(this.mediaDir, 'benchmedia01.1.m4a');
cpSync(mediaFiles.mp4, targetMp4);
cpSync(mediaFiles.m4a, targetM4a);
await seedDatabase(this.dbPath, { mp4: targetMp4, m4a: targetM4a });
}
async stop() {
if (this.serverProcess) {
this.serverProcess.kill('SIGKILL');
this.serverProcess = null;
await sleep(100);
}
}
async restartWithPublic(newPublicDir) {
await this.stop();
this.publicDir = newPublicDir;
await this.start();
}
}
/**
* Configure page for metrics collection
*/
async function setupPageMetrics(page) {
await page.addInitScript(() => {
window.__bootDone = null;
window.__fcp = null;
window.__lcp = null;
window.__longTasksTotal = 0;
// Observe boot completion landmark (#cards view set by renderList())
const markBoot = () => {
const cards = document.getElementById('cards');
if (cards && cards.dataset && cards.dataset.view) {
if (!window.__bootDone) window.__bootDone = performance.now();
return true;
}
return false;
};
const t = setInterval(() => {
if (markBoot()) clearInterval(t);
}, 5);
// Observe Long Tasks API (supported on Chromium)
try {
new PerformanceObserver((list) => {
for (const entry of list.getEntries()) {
window.__longTasksTotal += entry.duration;
}
}).observe({ type: 'longtask', buffered: true });
} catch {}
// Observe LCP
try {
new PerformanceObserver((list) => {
const entries = list.getEntries();
if (entries.length > 0) {
window.__lcp = entries[entries.length - 1].startTime;
}
}).observe({ type: 'largest-contentful-paint', buffered: true });
} catch {}
});
}
/**
* Run Scenario: Cold Load
*/
async function runColdScenario({ browserType, server, proxy, profile }) {
const isWebKit = browserType.name() === 'webkit';
const launchArgs = isWebKit ? [] : ['--autoplay-policy=no-user-gesture-required'];
const browser = await browserType.launch({
headless: true,
args: launchArgs,
});
const contextOptions = { proxy: { server: `http://127.0.0.1:${proxy.port}` } };
const context = await browser.newContext(contextOptions);
const page = await context.newPage();
// If Chromium, configure CDP Network conditions and 4x CPU throttle
if (!isWebKit) {
const cdp = await context.newCDPSession(page);
await cdp.send('Network.enable');
const isLossy = profile === 'lossy';
await cdp.send('Network.emulateNetworkConditions', {
offline: false,
latency: isLossy ? 250 : 150,
downloadThroughput: isLossy ? (1.0 * 1000 * 1000) / 8 : (1.6 * 1000 * 1000) / 8,
uploadThroughput: (750 * 1000) / 8,
connectionType: 'cellular4g',
});
await cdp.send('Emulation.setCPUThrottlingRate', { rate: 4 });
}
await setupPageMetrics(page);
proxy.resetStats();
proxy.startTracking();
const swInstallStart = Date.now();
const navUrl = `http://127.0.0.1:${server.port}/?v=benchmedia01`;
await page.goto(navUrl);
// Wait for boot landmark
await page.waitForFunction(() => window.__bootDone !== null, { timeout: 15000 });
// Check deterministic media playing-ready
let mediaReadyMs = null;
let autoplayBlocked = false;
try {
const mediaCheck = await page.waitForFunction(() => {
const v = document.querySelector('video') || document.querySelector('audio');
if (!v) return false;
return v.readyState >= 3 || (!v.paused && v.currentTime > 0);
}, { timeout: 8000 });
if (mediaCheck) {
const state = await page.evaluate(() => {
const v = document.querySelector('video') || document.querySelector('audio');
return { paused: v.paused, readyState: v.readyState };
});
autoplayBlocked = state.paused;
mediaReadyMs = await page.evaluate(() => performance.now());
}
} catch {
autoplayBlocked = true;
}
// Wait for initial Service Worker activation
await page.evaluate(async () => {
const reg = await navigator.serviceWorker.ready;
if (reg.active && reg.active.state === 'activated') return;
await new Promise((resolve) => {
const w = reg.active || reg.installing;
if (w.state === 'activated') return resolve();
w.addEventListener('statechange', () => {
if (w.state === 'activated') resolve();
});
});
});
const swInstallDuration = Date.now() - swInstallStart;
proxy.stopTracking();
const stats = proxy.getStats();
const pageMetrics = await page.evaluate(() => {
const nav = performance.getEntriesByType('navigation')[0];
const paint = performance.getEntriesByType('paint');
const fcp = paint.find((p) => p.name === 'first-contentful-paint');
return {
fcp: fcp ? Math.round(fcp.startTime) : null,
lcp: window.__lcp ? Math.round(window.__lcp) : null,
domContentLoaded: nav ? Math.round(nav.domContentLoadedEventEnd - nav.startTime) : null,
bootDone: window.__bootDone ? Math.round(window.__bootDone) : null,
longTaskTotal: Math.round(window.__longTasksTotal || 0),
};
});
const storage = await page.evaluate(async () => {
if (navigator.storage && navigator.storage.estimate) {
const est = await navigator.storage.estimate();
return { usage: est.usage || 0, quota: est.quota || 0 };
}
return { usage: null, quota: null };
});
await browser.close();
return {
requestCount: stats.requestCount,
wireBytes: stats.totalWireBytes,
wireBytesByType: stats.bytesByType,
fcp: pageMetrics.fcp,
lcp: pageMetrics.lcp,
domContentLoaded: pageMetrics.domContentLoaded,
bootDone: pageMetrics.bootDone,
longTaskTotal: pageMetrics.longTaskTotal,
swInstallDuration,
swInstallBytes: stats.totalWireBytes,
storageEstimate: storage,
mediaReady: mediaReadyMs ? Math.round(mediaReadyMs) : null,
autoplayBlocked,
};
}
/**
* Run Scenario: Warm Load (re-navigation with active SW)
*/
async function runWarmScenario({ browserType, server, proxy, profile }) {
const isWebKit = browserType.name() === 'webkit';
const launchArgs = isWebKit ? [] : ['--autoplay-policy=no-user-gesture-required'];
const browser = await browserType.launch({
headless: true,
args: launchArgs,
});
const contextOptions = { proxy: { server: `http://127.0.0.1:${proxy.port}` } };
const context = await browser.newContext(contextOptions);
const page = await context.newPage();
if (!isWebKit) {
const cdp = await context.newCDPSession(page);
await cdp.send('Network.enable');
const isLossy = profile === 'lossy';
await cdp.send('Network.emulateNetworkConditions', {
offline: false,
latency: isLossy ? 250 : 150,
downloadThroughput: isLossy ? (1.0 * 1000 * 1000) / 8 : (1.6 * 1000 * 1000) / 8,
uploadThroughput: (750 * 1000) / 8,
connectionType: 'cellular4g',
});
await cdp.send('Emulation.setCPUThrottlingRate', { rate: 4 });
}
await setupPageMetrics(page);
const navUrl = `http://127.0.0.1:${server.port}/`;
// Prime context with cold load
await page.goto(navUrl);
await page.waitForFunction(() => window.__bootDone !== null, { timeout: 15000 });
await page.evaluate(async () => {
const reg = await navigator.serviceWorker.ready;
if (reg.active && reg.active.state === 'activated') return;
await new Promise((resolve) => {
const w = reg.active || reg.installing;
if (w.state === 'activated') return resolve();
w.addEventListener('statechange', () => {
if (w.state === 'activated') resolve();
});
});
});
// Now measure warm re-navigation
proxy.resetStats();
proxy.startTracking();
await page.goto(navUrl);
await page.waitForFunction(() => window.__bootDone !== null, { timeout: 15000 });
proxy.stopTracking();
const stats = proxy.getStats();
const pageMetrics = await page.evaluate(() => {
const nav = performance.getEntriesByType('navigation')[0];
const paint = performance.getEntriesByType('paint');
const fcp = paint.find((p) => p.name === 'first-contentful-paint');
return {
fcp: fcp ? Math.round(fcp.startTime) : null,
lcp: window.__lcp ? Math.round(window.__lcp) : null,
domContentLoaded: nav ? Math.round(nav.domContentLoadedEventEnd - nav.startTime) : null,
bootDone: window.__bootDone ? Math.round(window.__bootDone) : null,
longTaskTotal: Math.round(window.__longTasksTotal || 0),
};
});
await browser.close();
return {
requestCount: stats.requestCount,
wireBytes: stats.totalWireBytes,
wireBytesByType: stats.bytesByType,
fcp: pageMetrics.fcp,
lcp: pageMetrics.lcp,
domContentLoaded: pageMetrics.domContentLoaded,
bootDone: pageMetrics.bootDone,
longTaskTotal: pageMetrics.longTaskTotal,
};
}
/**
* Run Scenario: Offline Reload under SW
*/
async function runOfflineScenario({ browserType, server, proxy, profile }) {
const browser = await browserType.launch({ headless: true });
const isWebKit = browserType.name() === 'webkit';
const contextOptions = { proxy: { server: `http://127.0.0.1:${proxy.port}` } };
const context = await browser.newContext(contextOptions);
const page = await context.newPage();
await setupPageMetrics(page);
const navUrl = `http://127.0.0.1:${server.port}/`;
// Prime SW
await page.goto(navUrl);
await page.waitForFunction(() => window.__bootDone !== null, { timeout: 15000 });
await page.evaluate(async () => {
const reg = await navigator.serviceWorker.ready;
if (reg.active && reg.active.state === 'activated') return;
await new Promise((resolve) => {
const w = reg.active || reg.installing;
if (w.state === 'activated') return resolve();
w.addEventListener('statechange', () => {
if (w.state === 'activated') resolve();
});
});
});
// Switch to offline mode:
// On Chromium, context.setOffline(true) works.
// On Linux WebKit, Playwright context.setOffline throws an internal browser error,
// so proxy.setOffline(true) cuts upstream network to simulate offline truthfully.
if (!isWebKit) {
await context.setOffline(true);
}
proxy.setOffline(true);
proxy.resetStats();
proxy.startTracking();
const t0 = Date.now();
await page.reload();
await page.waitForFunction(() => window.__bootDone !== null, { timeout: 15000 });
const reloadDuration = Date.now() - t0;
proxy.stopTracking();
proxy.setOffline(false);
const stats = proxy.getStats();
const pageMetrics = await page.evaluate(() => {
const nav = performance.getEntriesByType('navigation')[0];
const paint = performance.getEntriesByType('paint');
const fcp = paint.find((p) => p.name === 'first-contentful-paint');
return {
fcp: fcp ? Math.round(fcp.startTime) : null,
lcp: window.__lcp ? Math.round(window.__lcp) : null,
domContentLoaded: nav ? Math.round(nav.domContentLoadedEventEnd - nav.startTime) : null,
bootDone: window.__bootDone ? Math.round(window.__bootDone) : null,
};
});
await browser.close();
return {
requestCount: stats.requestCount,
wireBytes: stats.totalWireBytes,
fcp: pageMetrics.fcp,
lcp: pageMetrics.lcp,
domContentLoaded: pageMetrics.domContentLoaded,
bootDone: pageMetrics.bootDone,
reloadDuration,
};
}
/**
* Run Scenario: Update N -> N+1
*/
async function runUpdateScenario({ browserType, server, proxy, profile, updateKind, trees }) {
const browser = await browserType.launch({ headless: true });
const isWebKit = browserType.name() === 'webkit';
const contextOptions = { proxy: { server: `http://127.0.0.1:${proxy.port}` } };
const context = await browser.newContext(contextOptions);
const page = await context.newPage();
if (!isWebKit) {
const cdp = await context.newCDPSession(page);
await cdp.send('Network.enable');
const isLossy = profile === 'lossy';
await cdp.send('Network.emulateNetworkConditions', {
offline: false,
latency: isLossy ? 250 : 150,
downloadThroughput: isLossy ? (1.0 * 1000 * 1000) / 8 : (1.6 * 1000 * 1000) / 8,
uploadThroughput: (750 * 1000) / 8,
connectionType: 'cellular4g',
});
}
// Start on Tree N (base)
await server.restartWithPublic(trees.treeBase);
const navUrl = `http://127.0.0.1:${server.port}/`;
await page.goto(navUrl);
// Wait for steady state: active worker present, no waiting worker
while (true) {
const ready = await page.evaluate(async () => {
const reg = await navigator.serviceWorker.getRegistration();
return reg?.active?.state === 'activated' && !reg?.installing && !reg?.waiting;
});
if (ready) break;
await sleep(50);
}
// Switch server to Tree N+1 (app.js or theme-glass.css modification)
const targetTree = updateKind === 'js' ? trees.treeUpdateJs : trees.treeUpdateCss;
await server.restartWithPublic(targetTree);
// Measure wire bytes and duration from reg.update() until new SW is waiting/installed
proxy.resetStats();
proxy.startTracking();
const t0 = Date.now();
await page.evaluate(() => navigator.serviceWorker.getRegistration().then((r) => r.update()));
// Wait until new worker reaches waiting state
while (true) {
const waiting = await page.evaluate(async () => {
const reg = await navigator.serviceWorker.getRegistration();
return reg?.waiting?.state === 'installed';
});
if (waiting) break;
await sleep(50);
}
const duration = Date.now() - t0;
proxy.stopTracking();
const stats = proxy.getStats();
await browser.close();
return {
updateKind,
duration,
requestCount: stats.requestCount,
wireBytes: stats.totalWireBytes,
wireBytesByType: stats.bytesByType,
};
}
/**
* Main execution
*/
async function main() {
const options = parseArgs();
console.log('=== YTPlayer Performance Baseline Harness ===');
console.log(`Runs per scenario: ${options.runs}`);
console.log(`Browser target: ${options.browser}`);
console.log(`Network profile: ${options.profile}`);
console.log(`Scenario target: ${options.scenario}`);
const browsers = [];
if (options.browser === 'all' || options.browser === 'chromium') browsers.push(pw.chromium);
if (options.browser === 'all' || options.browser === 'webkit') browsers.push(pw.webkit);
const profiles = [];
if (options.profile === 'all' || options.profile === 'lte') profiles.push('lte');
if (options.profile === 'all' || options.profile === 'lossy') profiles.push('lossy');
const tmpRoot = path.join(REPO_ROOT, 'perf', '.tmp', `bench-${Date.now()}`);
mkdirSync(tmpRoot, { recursive: true });
// Setup scratch trees for N -> N+1 update scenarios
const treeBase = path.join(tmpRoot, 'tree-base');
const treeUpdateJs = path.join(tmpRoot, 'tree-update-js');
const treeUpdateCss = path.join(tmpRoot, 'tree-update-css');
cpSync(path.join(REPO_ROOT, 'frontend'), treeBase, { recursive: true });
cpSync(path.join(REPO_ROOT, 'frontend'), treeUpdateJs, { recursive: true });
cpSync(path.join(REPO_ROOT, 'frontend'), treeUpdateCss, { recursive: true });
// Append 1 comment line to app.js in treeUpdateJs
appendFileSync(path.join(treeUpdateJs, 'app.js'), '\n// baseline bench: 1-line app.js comment\n');
// Append 1 rule to theme-glass.css in treeUpdateCss
appendFileSync(path.join(treeUpdateCss, 'theme-glass.css'), '\n/* baseline bench: 1-line css rule */\n.baseline-marker { opacity: 0.99; }\n');
const trees = { treeBase, treeUpdateJs, treeUpdateCss };
const serverScratch = path.join(tmpRoot, 'server-run');
const server = new ServerManager(serverScratch, treeBase);
await server.start();
console.log(`Bun server running on ephemeral port ${server.port}`);
const results = {
date: new Date().toISOString(),
commit: spawnSync('git', ['rev-parse', 'HEAD'], { cwd: REPO_ROOT }).stdout.toString().trim(),
runsConfigured: options.runs,
browsers: {},
};
try {
for (const bType of browsers) {
const bName = bType.name();
results.browsers[bName] = {};
for (const prof of profiles) {
console.log(`\n>>> Running Benchmark: Browser=[${bName}] Profile=[${prof}]`);
results.browsers[bName][prof] = {};
const proxyProfile = bName === 'chromium' ? 'unthrottled' : prof;
const proxy = createThrottleProxy({ targetPort: server.port, profile: proxyProfile });
proxy.port = await proxy.listen();
console.log(`Proxy listening on port ${proxy.port} (mode: ${proxyProfile})`);
// 1. Cold Scenario
if (options.scenario === 'all' || options.scenario === 'cold') {
console.log(` [Scenario: Cold Load] running ${options.runs} iterations...`);
const coldRuns = [];
for (let r = 0; r < options.runs; r++) {
process.stdout.write(` run ${r + 1}/${options.runs}... `);
const res = await runColdScenario({ browserType: bType, server, proxy, profile: prof });
coldRuns.push(res);
console.log(`done (${res.wireBytes} bytes, ${res.requestCount} reqs, boot=${res.bootDone}ms)`);
}
results.browsers[bName][prof].cold = {
requestCount: summarizeList(coldRuns.map((r) => r.requestCount)),
wireBytes: summarizeList(coldRuns.map((r) => r.wireBytes)),
fcp: summarizeList(coldRuns.map((r) => r.fcp)),
lcp: summarizeList(coldRuns.map((r) => r.lcp)),
domContentLoaded: summarizeList(coldRuns.map((r) => r.domContentLoaded)),
bootDone: summarizeList(coldRuns.map((r) => r.bootDone)),
longTaskTotal: summarizeList(coldRuns.map((r) => r.longTaskTotal)),
swInstallDuration: summarizeList(coldRuns.map((r) => r.swInstallDuration)),
swInstallBytes: summarizeList(coldRuns.map((r) => r.swInstallBytes)),
storageEstimate: coldRuns[0]?.storageEstimate || null,
mediaReady: summarizeList(coldRuns.map((r) => r.mediaReady)),
autoplayBlocked: coldRuns.some((r) => r.autoplayBlocked),
wireBytesByType: coldRuns[0]?.wireBytesByType || {},
};
}
// 2. Warm Scenario
if (options.scenario === 'all' || options.scenario === 'warm') {
console.log(` [Scenario: Warm Load] running ${options.runs} iterations...`);
const warmRuns = [];
for (let r = 0; r < options.runs; r++) {
process.stdout.write(` run ${r + 1}/${options.runs}... `);
const res = await runWarmScenario({ browserType: bType, server, proxy, profile: prof });
warmRuns.push(res);
console.log(`done (${res.wireBytes} bytes, ${res.requestCount} reqs, boot=${res.bootDone}ms)`);
}
results.browsers[bName][prof].warm = {
requestCount: summarizeList(warmRuns.map((r) => r.requestCount)),
wireBytes: summarizeList(warmRuns.map((r) => r.wireBytes)),
fcp: summarizeList(warmRuns.map((r) => r.fcp)),
lcp: summarizeList(warmRuns.map((r) => r.lcp)),
domContentLoaded: summarizeList(warmRuns.map((r) => r.domContentLoaded)),
bootDone: summarizeList(warmRuns.map((r) => r.bootDone)),
longTaskTotal: summarizeList(warmRuns.map((r) => r.longTaskTotal)),
};
}
// 3. Offline Scenario
if (options.scenario === 'all' || options.scenario === 'offline') {
console.log(` [Scenario: Offline Reload] running ${options.runs} iterations...`);
const offRuns = [];
for (let r = 0; r < options.runs; r++) {
process.stdout.write(` run ${r + 1}/${options.runs}... `);
const res = await runOfflineScenario({ browserType: bType, server, proxy, profile: prof });
offRuns.push(res);
console.log(`done (reload=${res.reloadDuration}ms, boot=${res.bootDone}ms)`);
}
results.browsers[bName][prof].offline = {
requestCount: summarizeList(offRuns.map((r) => r.requestCount)),
wireBytes: summarizeList(offRuns.map((r) => r.wireBytes)),
fcp: summarizeList(offRuns.map((r) => r.fcp)),
lcp: summarizeList(offRuns.map((r) => r.lcp)),
domContentLoaded: summarizeList(offRuns.map((r) => r.domContentLoaded)),
bootDone: summarizeList(offRuns.map((r) => r.bootDone)),
reloadDuration: summarizeList(offRuns.map((r) => r.reloadDuration)),
};
}
// 4. Update N -> N+1 (JS)
if (options.scenario === 'all' || options.scenario === 'update-js') {
console.log(` [Scenario: Update N->N+1 (1-line app.js)] running ${options.runs} iterations...`);
const updJsRuns = [];
for (let r = 0; r < options.runs; r++) {
process.stdout.write(` run ${r + 1}/${options.runs}... `);
const res = await runUpdateScenario({
browserType: bType,
server,
proxy,
profile: prof,
updateKind: 'js',
trees,
});
updJsRuns.push(res);
console.log(`done (${res.wireBytes} bytes, ${res.requestCount} reqs, dur=${res.duration}ms)`);
}
results.browsers[bName][prof].updateJs = {
requestCount: summarizeList(updJsRuns.map((r) => r.requestCount)),
wireBytes: summarizeList(updJsRuns.map((r) => r.wireBytes)),
duration: summarizeList(updJsRuns.map((r) => r.duration)),
wireBytesByType: updJsRuns[0]?.wireBytesByType || {},
};
}
// 5. Update N -> N+1 (CSS)
if (options.scenario === 'all' || options.scenario === 'update-css') {
console.log(` [Scenario: Update N->N+1 (1-line theme-glass.css)] running ${options.runs} iterations...`);
const updCssRuns = [];
for (let r = 0; r < options.runs; r++) {
process.stdout.write(` run ${r + 1}/${options.runs}... `);
const res = await runUpdateScenario({
browserType: bType,
server,
proxy,
profile: prof,
updateKind: 'css',
trees,
});
updCssRuns.push(res);
console.log(`done (${res.wireBytes} bytes, ${res.requestCount} reqs, dur=${res.duration}ms)`);
}
results.browsers[bName][prof].updateCss = {
requestCount: summarizeList(updCssRuns.map((r) => r.requestCount)),
wireBytes: summarizeList(updCssRuns.map((r) => r.wireBytes)),
duration: summarizeList(updCssRuns.map((r) => r.duration)),
wireBytesByType: updCssRuns[0]?.wireBytesByType || {},
};
}
await proxy.close();
}
}
} finally {
await server.stop();
rmSync(tmpRoot, { recursive: true, force: true });
}
// Write results JSON
const now = new Date();
const dateStr = `${now.getFullYear()}-${String(now.getMonth() + 1).padStart(2, '0')}-${String(now.getDate()).padStart(2, '0')}`;
const outPath = options.out || path.join(REPO_ROOT, 'perf', 'results', `baseline-${dateStr}.json`);
mkdirSync(path.dirname(outPath), { recursive: true });
writeFileSync(outPath, JSON.stringify(results, null, 2));
console.log(`\nBaseline results saved to: ${outPath}`);
// Print Summary Table
printSummaryTable(results);
// Compare mode if requested
if (options.compare) {
compareResults(options.compare, results);
}
}
function printSummaryTable(results) {
console.log('\n============================= HEADLINE BASELINE SUMMARY =============================');
for (const [browser, profiles] of Object.entries(results.browsers)) {
for (const [profile, scenarios] of Object.entries(profiles)) {
console.log(`\n[Browser: ${browser} | Profile: ${profile}]`);
console.log('| Scenario | Req Count | Wire Bytes | Boot / Ready | SW Dur / Reload |');
console.log('|---------------------------|-----------|------------|--------------|-----------------|');
if (scenarios.cold) {
const c = scenarios.cold;
console.log(
`| Cold First Load | ${String(c.requestCount.median).padEnd(9)} | ${formatBytes(c.wireBytes.median).padEnd(10)} | ${String(c.bootDone.median + 'ms').padEnd(12)} | ${String(c.swInstallDuration.median + 'ms').padEnd(15)} |`
);
}
if (scenarios.warm) {
const w = scenarios.warm;
console.log(
`| Warm Reload (under SW) | ${String(w.requestCount.median).padEnd(9)} | ${formatBytes(w.wireBytes.median).padEnd(10)} | ${String(w.bootDone.median + 'ms').padEnd(12)} | ${'-'.padEnd(15)} |`
);
}
if (scenarios.offline) {
const o = scenarios.offline;
console.log(
`| Offline Reload (under SW) | ${String(o.requestCount.median).padEnd(9)} | ${formatBytes(o.wireBytes.median).padEnd(10)} | ${String(o.bootDone.median + 'ms').padEnd(12)} | ${String(o.reloadDuration.median + 'ms').padEnd(15)} |`
);
}
if (scenarios.updateJs) {
const u = scenarios.updateJs;
console.log(
`| Update N->N+1 (1-line JS) | ${String(u.requestCount.median).padEnd(9)} | ${formatBytes(u.wireBytes.median).padEnd(10)} | ${'-'.padEnd(12)} | ${String(u.duration.median + 'ms').padEnd(15)} |`
);
}
if (scenarios.updateCss) {
const u = scenarios.updateCss;
console.log(
`| Update N->N+1 (1-line CSS)| ${String(u.requestCount.median).padEnd(9)} | ${formatBytes(u.wireBytes.median).padEnd(10)} | ${'-'.padEnd(12)} | ${String(u.duration.median + 'ms').padEnd(15)} |`
);
}
}
}
console.log('\n=====================================================================================');
}
function formatBytes(bytes) {
if (bytes === null || bytes === undefined) return 'N/A';
if (bytes >= 1024 * 1024) return (bytes / (1024 * 1024)).toFixed(2) + ' MB';
if (bytes >= 1024) return (bytes / 1024).toFixed(1) + ' KB';
return bytes + ' B';
}
function compareResults(baselinePath, currentResults) {
console.log(`\n================ Phase 5 Comparison against: ${baselinePath} ================`);
if (!existsSync(baselinePath)) {
console.error(`Error: Baseline file not found: ${baselinePath}`);
return;
}
const base = JSON.parse(readFileSync(baselinePath, 'utf8'));
for (const [browser, profiles] of Object.entries(currentResults.browsers)) {
for (const [profile, scenarios] of Object.entries(profiles)) {
console.log(`\n[Comparison: ${browser} / ${profile}]`);
console.log('| Metric | Baseline | Current | Delta (%) |');
console.log('|------------------------------------|-------------|-------------|--------------|');
const baseScen = base.browsers?.[browser]?.[profile];
if (!baseScen) {
console.log('| (No matching baseline data) | | | |');
continue;
}
const rows = [
['Cold: Wire Bytes', baseScen.cold?.wireBytes?.median, scenarios.cold?.wireBytes?.median, true],
['Cold: Boot Done (ms)', baseScen.cold?.bootDone?.median, scenarios.cold?.bootDone?.median, false],
['Cold: Requests', baseScen.cold?.requestCount?.median, scenarios.cold?.requestCount?.median, false],
['Warm: Boot Done (ms)', baseScen.warm?.bootDone?.median, scenarios.warm?.bootDone?.median, false],
['Update (1-line JS): Wire Bytes', baseScen.updateJs?.wireBytes?.median, scenarios.updateJs?.wireBytes?.median, true],
['Update (1-line JS): Time (ms)', baseScen.updateJs?.duration?.median, scenarios.updateJs?.duration?.median, false],
['Update (1-line CSS): Wire Bytes', baseScen.updateCss?.wireBytes?.median, scenarios.updateCss?.wireBytes?.median, true],
['Update (1-line CSS): Time (ms)', baseScen.updateCss?.duration?.median, scenarios.updateCss?.duration?.median, false],
];
for (const [name, bVal, cVal, isBytes] of rows) {
if (bVal === undefined || cVal === undefined) continue;
const bStr = isBytes ? formatBytes(bVal) : bVal !== null ? `${bVal}` : 'N/A';
const cStr = isBytes ? formatBytes(cVal) : cVal !== null ? `${cVal}` : 'N/A';
let deltaStr = '-';
if (typeof bVal === 'number' && typeof cVal === 'number' && bVal > 0) {
const pct = (((cVal - bVal) / bVal) * 100).toFixed(1);
const sign = cVal >= bVal ? '+' : '';
deltaStr = `${sign}${pct}%`;
}
console.log(`| ${name.padEnd(34)} | ${bStr.padEnd(11)} | ${cStr.padEnd(11)} | ${deltaStr.padEnd(12)} |`);
}
}
}
console.log('==============================================================================');
}
main().catch((err) => {
console.error('Benchmark fatal error:', err);
process.exit(1);
});

126
perf/make-shell-fixture.sh Executable file
View File

@@ -0,0 +1,126 @@
#!/usr/bin/env bash
set -euo pipefail
# ==============================================================================
# perf/make-shell-fixture.sh
# Archives frontend from <commit> into perf/fixtures/shell-<tag>/,
# computes the build tag the way server.js computeBuildTag() does,
# and verifies that it matches expected-tag (or live prod tag).
# ==============================================================================
if [ $# -lt 1 ]; then
echo "Usage: $0 <commit> [expected-tag]" >&2
exit 1
fi
COMMIT="$1"
EXPECTED_TAG="${2:-}"
REPO_ROOT="$(cd "$(dirname "$0")/.." && pwd)"
PERF_DIR="$REPO_ROOT/perf"
FIXTURES_DIR="$PERF_DIR/fixtures"
TMP_DIR="$PERF_DIR/.tmp"
mkdir -p "$FIXTURES_DIR" "$TMP_DIR"
PROD_URL="https://worship.hesed.sbs/api/version"
fetch_prod_tag() {
curl -s -S --max-time 10 "$PROD_URL" | node -e '
let data = "";
process.stdin.on("data", chunk => { data += chunk; });
process.stdin.on("end", () => {
try {
const j = JSON.parse(data);
if (j.buildTag) {
process.stdout.write(j.buildTag);
process.exit(0);
}
} catch {}
process.exit(1);
});
' 2>/dev/null || echo ""
}
# Verify the commit exists in git
if ! git rev-parse --verify "$COMMIT^{commit}" >/dev/null 2>&1; then
echo "Error: commit '$COMMIT' not found in git repository" >&2
exit 1
fi
RESOLVED_COMMIT="$(git rev-parse --short=7 "$COMMIT")"
echo "Extracting frontend from commit $RESOLVED_COMMIT..."
STAGING_DIR="$(mktemp -d "$TMP_DIR/fixture-staging-XXXXXX")"
cleanup() {
rm -rf "$STAGING_DIR"
}
trap cleanup EXIT
# Archive frontend tree into staging dir
git archive "$COMMIT" frontend | tar -x -C "$STAGING_DIR" --strip-components=1
# Compute build tag using the exact logic from server/server.js computeBuildTag()
COMPUTED_TAG=$(node -e '
const { createHash } = require("crypto");
const { readdirSync, statSync, readFileSync } = require("fs");
function computeTag(dir) {
const hash = createHash("sha256");
const walk = (d, prefix) => {
for (const name of readdirSync(d).sort()) {
const full = `${d}/${name}`;
const rel = `${prefix}/${name}`;
if (statSync(full).isDirectory()) {
walk(full, rel);
} else {
hash.update(rel);
hash.update(readFileSync(full));
}
}
};
walk(dir, "./public");
return hash.digest("hex").slice(0, 12);
}
const tag = computeTag(process.argv[1]);
process.stdout.write(tag);
' "$STAGING_DIR")
echo "Computed build tag for $RESOLVED_COMMIT: $COMPUTED_TAG"
# Determine target expected tag
if [ -z "$EXPECTED_TAG" ]; then
echo "Querying live production tag from $PROD_URL..."
EXPECTED_TAG="$(fetch_prod_tag)"
if [ -z "$EXPECTED_TAG" ]; then
echo "Error: failed to fetch live buildTag from $PROD_URL" >&2
exit 1
fi
echo "Live production build tag: $EXPECTED_TAG"
# If mismatch, deploy might be in progress: wait 5 minutes and retry once
if [ "$COMPUTED_TAG" != "$EXPECTED_TAG" ]; then
echo "Tag mismatch detected (computed=$COMPUTED_TAG vs prod=$EXPECTED_TAG)."
echo "Deploy may be in progress. Waiting 300 seconds to retry once..."
sleep 300
EXPECTED_TAG="$(fetch_prod_tag)"
echo "Retried production build tag: $EXPECTED_TAG"
fi
fi
if [ "$COMPUTED_TAG" != "$EXPECTED_TAG" ]; then
echo "Error: build tag mismatch! Computed '$COMPUTED_TAG' does not match expected '$EXPECTED_TAG'" >&2
exit 1
fi
TARGET_FIXTURE_DIR="$FIXTURES_DIR/shell-$COMPUTED_TAG"
echo "Installing fixture into $TARGET_FIXTURE_DIR..."
rm -rf "$TARGET_FIXTURE_DIR"
mkdir -p "$TARGET_FIXTURE_DIR"
cp -a "$STAGING_DIR"/. "$TARGET_FIXTURE_DIR"/
FILE_COUNT=$(find "$TARGET_FIXTURE_DIR" -type f | wc -l)
echo "Success: shell fixture created at $TARGET_FIXTURE_DIR with $FILE_COUNT files."
echo "Verified commit: $RESOLVED_COMMIT"
echo "Verified build tag: $COMPUTED_TAG"

201
perf/proxy.mjs Normal file
View File

@@ -0,0 +1,201 @@
import http from 'node:http';
import { Transform } from 'node:stream';
/**
* ThrottleTransform
* Delays stream chunks to simulate bandwidth caps (e.g. 1.6 Mbps = 200,000 B/s).
*/
class ThrottleTransform extends Transform {
constructor(bytesPerSec) {
super();
this.bytesPerSec = bytesPerSec;
}
_transform(chunk, encoding, callback) {
if (!this.bytesPerSec || this.bytesPerSec <= 0) {
this.push(chunk);
return callback();
}
const delayMs = (chunk.length / this.bytesPerSec) * 1000;
setTimeout(() => {
this.push(chunk);
callback();
}, Math.min(delayMs, 2000));
}
}
/**
* Categorize Content-Type into standard metric buckets.
*/
export function categorizeContentType(contentType, urlPath) {
const ct = (contentType || '').toLowerCase();
const path = (urlPath || '').toLowerCase();
if (ct.includes('text/html') || path.endsWith('.html')) return 'html';
if (ct.includes('javascript') || path.endsWith('.js') || path.endsWith('.mjs')) return 'js';
if (ct.includes('text/css') || path.endsWith('.css')) return 'css';
if (ct.includes('font') || path.endsWith('.woff2') || path.endsWith('.woff') || path.endsWith('.ttf')) return 'fonts';
if (ct.includes('image/') || path.endsWith('.png') || path.endsWith('.jpg') || path.endsWith('.svg') || path.endsWith('.webp') || path.endsWith('.ico')) return 'images';
if (ct.includes('video/') || ct.includes('audio/') || path.endsWith('.mp4') || path.endsWith('.m4a') || path.endsWith('.mp3')) return 'media';
if (ct.includes('application/json') || path.startsWith('/api/')) return 'api';
return 'other';
}
/**
* Creates an HTTP throttling & stalling proxy for WebKit (and browser network tracing).
* Supports both forward proxying and reverse proxying to targetPort.
*/
export function createThrottleProxy({ targetPort, profile = 'lte' }) {
let activeTracking = false;
let requestCount = 0;
let totalWireBytes = 0;
let bytesByType = {
html: 0,
js: 0,
css: 0,
fonts: 0,
images: 0,
media: 0,
api: 0,
other: 0,
};
let reqIndex = 0;
let isOfflineMode = false;
// Profile configuration:
// LTE: 150ms roundtrip latency, 1.6 Mbps (200 KB/s) download, 750 kbps upload
// Lossy: 250ms latency, 1.0 Mbps (125 KB/s) download, stall every 7th request by 1500ms
// Unthrottled: 0 latency, 0 throttle (for Chromium where CDP throttles network/CPU)
const isLossy = profile === 'lossy';
const isUnthrottled = profile === 'unthrottled';
const latencyMs = isUnthrottled ? 0 : (isLossy ? 250 : 150);
const downloadBytesPerSec = isUnthrottled ? 0 : (isLossy ? (1.0 * 1000 * 1000) / 8 : (1.6 * 1000 * 1000) / 8);
const server = http.createServer((clientReq, clientRes) => {
if (isOfflineMode) {
clientReq.destroy();
return;
}
reqIndex++;
const currentReqIndex = reqIndex;
let stallMs = 0;
if (isLossy && currentReqIndex % 7 === 0) {
stallMs = 1500; // Simulated lossy stall
}
const totalDelay = isUnthrottled ? 0 : (latencyMs + stallMs);
const rawUrl = clientReq.url;
const targetUrl = rawUrl.startsWith('http://') || rawUrl.startsWith('https://')
? rawUrl
: `http://127.0.0.1:${targetPort}${rawUrl}`;
const parsedTarget = new URL(targetUrl);
const pathname = parsedTarget.pathname;
const options = {
protocol: parsedTarget.protocol,
hostname: parsedTarget.hostname,
port: parsedTarget.port,
path: parsedTarget.pathname + parsedTarget.search,
method: clientReq.method,
headers: { ...clientReq.headers, host: parsedTarget.host },
};
const forward = () => {
const upReq = http.request(options, (upRes) => {
const ct = upRes.headers['content-type'] || '';
const cat = categorizeContentType(ct, pathname);
if (activeTracking) {
requestCount++;
}
clientRes.writeHead(upRes.statusCode, upRes.headers);
if (isUnthrottled) {
upRes.on('data', (chunk) => {
if (activeTracking) {
totalWireBytes += chunk.length;
bytesByType[cat] = (bytesByType[cat] || 0) + chunk.length;
}
clientRes.write(chunk);
});
upRes.on('end', () => clientRes.end());
} else {
const throttler = new ThrottleTransform(downloadBytesPerSec);
throttler.on('data', (chunk) => {
if (activeTracking) {
totalWireBytes += chunk.length;
bytesByType[cat] = (bytesByType[cat] || 0) + chunk.length;
}
});
upRes.pipe(throttler).pipe(clientRes);
}
});
upReq.on('error', (err) => {
try {
clientRes.statusCode = 502;
clientRes.end(err.message);
} catch {}
});
clientReq.pipe(upReq);
};
if (totalDelay > 0) {
setTimeout(forward, totalDelay);
} else {
forward();
}
});
return {
server,
listen: () =>
new Promise((resolve) => {
server.listen(0, '127.0.0.1', () => {
const port = server.address().port;
resolve(port);
});
}),
close: () =>
new Promise((resolve) => {
server.close(resolve);
}),
startTracking: () => {
activeTracking = true;
},
stopTracking: () => {
activeTracking = false;
},
resetStats: () => {
requestCount = 0;
totalWireBytes = 0;
bytesByType = {
html: 0,
js: 0,
css: 0,
fonts: 0,
images: 0,
media: 0,
api: 0,
other: 0,
};
},
setOffline: (val) => {
isOfflineMode = !!val;
},
getStats: () => ({
requestCount,
totalWireBytes,
bytesByType: { ...bytesByType },
}),
};
}

1639
perf/results/baseline-2026-10-07.json Normal file
View File

File diff suppressed because it is too large Load Diff

175
plans/perf-baseline.md Normal file
View File

@@ -0,0 +1,175 @@
# Performance Baseline Report (Phase 0 / UNCHANGED code)
**Date**: 2026-10-07
**Commit**: `b77938a`
**Dataset**: `perf/results/baseline-2026-10-07.json`
---
## 1. Executive Summary & Headline Baseline Numbers
This baseline captures the performance metrics of the unoptimized ytplayer codebase (commit `b77938a`, matching production tag `1175f1a1d2c1`) across 5 scenarios on both Chromium and WebKit under simulated LTE and Lossy network profiles (median of 5 runs).
### Headline Baseline Numbers
- **Cold First-Load Wire Transfer**:
- **Chromium**: 164 requests, **1,015.0 KB** compressed wire bytes (boot-done median: **10,163 ms** under 4× CPU throttle + LTE).
- **WebKit**: 162 requests, **1,152.0 KB (1.12 MB)** wire bytes (boot-done median: **3,201 ms** under proxy LTE).
- Breakdown: ~477 KB JavaScript, ~397 KB Web Fonts, ~131 KB CSS, ~27 KB HTML.
- **Warm Reload (under Service Worker)**:
- **Chromium**: 4 requests, **5.2 KB** wire bytes (periodic API version poll), boot-done median: **675 ms** (LTE) / **651 ms** (Lossy).
- **WebKit**: 0 requests, **0 KB** wire bytes, boot-done median: **131 ms** (LTE) / **131 ms** (Lossy).
- **Offline Reload (under Service Worker)**:
- **Chromium**: 0 requests, **0 B** wire bytes, reload median: **240 ms**, boot-done median: **219 ms**.
- **WebKit**: 0 requests, **0 B** wire bytes, reload median: **135 ms**, boot-done median: **128 ms**.
- **Update N -> N+1 Wire Cost (The Invalidation Penalty)**:
- **1-line `app.js` change**:
- **Chromium**: 88 requests, **596.6 KB** wire bytes (~611 KB uncompressed/brotli), duration: **2,013 ms** (LTE) / **2,031 ms** (Lossy).
- **WebKit**: 86 requests, **664.3 KB** wire bytes (~680 KB wire), duration: **4,457 ms** (LTE) / **8,135 ms** (Lossy).
- **1-line `theme-glass.css` change**:
- **Chromium**: 88 requests, **596.7 KB** wire bytes, duration: **1,970 ms** (LTE) / **2,004 ms** (Lossy).
- **WebKit**: 86-87 requests, **664.3 KB** wire bytes, duration: **4,510 ms** (LTE) / **8,269 ms** (Lossy).
> **Core Invalidation Finding**: A single 1-line change to *any* static file (`app.js` or `theme-glass.css`) mutates `BUILD_TAG` calculated by `computeBuildTag()`. This causes `sw.js` to re-fetch and re-cache all 86-88 shell assets (~600-664 KB wire payload) rather than just the mutated resource.
---
## 2. Benchmark Harness Execution
The benchmark harness is located in `perf/baseline.mjs` and driven by Playwright with an isolated ephemeral Bun server and a custom pacing proxy (`perf/proxy.mjs`).
### Quick Start / Exact Commands
```bash
# Ensure node_modules is symlinked to root
ln -s ~/development/personal/ytplayer/node_modules node_modules
# Run full baseline suite (all scenarios, both browsers, both profiles, 5 runs)
node perf/baseline.mjs --runs 5 --browser all --profile all --scenario all --out perf/results/baseline-2026-10-07.json
# Run individual browser or scenario
node perf/baseline.mjs --runs 5 --browser chromium --profile lte --scenario cold
node perf/baseline.mjs --runs 5 --browser webkit --profile lte --scenario update-js
# Run comparison against baseline after optimizations
node perf/baseline.mjs --runs 5 --browser all --profile all --scenario all --compare perf/results/baseline-2026-10-07.json
```
---
## 3. Results by Browser and Profile (Median of 5 Runs)
### Chromium — LTE (1.6 Mbps down, 750 kbps up, 150 ms RTT, 4× CPU throttle)
| Scenario | Reqs (med) | Wire Bytes (med) | FCP (ms) | LCP (ms) | Boot Done (ms) | SW Ready / Reload (ms) | Long Task Total (ms) |
|---|---|---|---|---|---|---|---|
| **Cold First Load** | 164 | 1,015.0 KB | 1,540 | 1,896 | 10,163 | 38,446 | 397 |
| **Warm Reload (SW)** | 4 | 5.2 KB | 180 | 684 | 675 | - | 156 |
| **Offline Reload** | 0 | 0.0 KB | 72 | 108 | 219 | 240 | - |
| **Update N->N+1 (JS)** | 88 | 596.6 KB | - | - | - | 2,013 | - |
| **Update N->N+1 (CSS)** | 88 | 596.7 KB | - | - | - | 1,970 | - |
- **Storage Estimate**: Quota = 4,298,447,325 B (~4.0 GB), Usage = 3,480,029 B (~3.31 MB).
- **Cold Wire Bytes by Type**: JS 477.2 KB, Fonts 397.1 KB, CSS 130.7 KB, HTML 27.1 KB, Images 6.3 KB, API/Other 1.0 KB.
---
### Chromium — Lossy (1.0 Mbps down, 500 kbps up, 250 ms RTT, 4× CPU throttle)
| Scenario | Reqs (med) | Wire Bytes (med) | FCP (ms) | LCP (ms) | Boot Done (ms) | SW Ready / Reload (ms) | Long Task Total (ms) |
|---|---|---|---|---|---|---|---|
| **Cold First Load** | 164 | 1,015.0 KB | 2,344 | 2,912 | 16,015 | 42,518 | 432 |
| **Warm Reload (SW)** | 4 | 5.2 KB | 184 | 644 | 651 | - | 155 |
| **Offline Reload** | 0 | 0.0 KB | 80 | 92 | 237 | 253 | - |
| **Update N->N+1 (JS)** | 88 | 596.6 KB | - | - | - | 2,031 | - |
| **Update N->N+1 (CSS)** | 88 | 596.7 KB | - | - | - | 2,004 | - |
---
### WebKit — LTE (1.6 Mbps down, 150 ms RTT via proxy)
| Scenario | Reqs (med) | Wire Bytes (med) | FCP (ms) | LCP (ms) | Boot Done (ms) | SW Ready / Reload (ms) | Long Task Total (ms) |
|---|---|---|---|---|---|---|---|
| **Cold First Load** | 162 | 1.12 MB | 1,088 | 1,287 | 3,201 | 30,508 | 0* |
| **Warm Reload (SW)** | 0 | 0.0 KB | 110 | 110 | 131 | - | 0* |
| **Offline Reload** | 0 | 0.0 KB | 107 | 107 | 128 | 135 | - |
| **Update N->N+1 (JS)** | 86 | 664.3 KB | - | - | - | 4,457 | - |
| **Update N->N+1 (CSS)** | 86 | 664.3 KB | - | - | - | 4,510 | - |
---
### WebKit — Lossy (1.0 Mbps down, 250 ms RTT + 1,500 ms stalls via proxy)
| Scenario | Reqs (med) | Wire Bytes (med) | FCP (ms) | LCP (ms) | Boot Done (ms) | SW Ready / Reload (ms) | Long Task Total (ms) |
|---|---|---|---|---|---|---|---|
| **Cold First Load** | 162 | 1.12 MB | 3,557 | 3,735 | 7,303 | 35,449 | 0* |
| **Warm Reload (SW)** | 0 | 0.0 KB | 111 | 112 | 131 | - | 0* |
| **Offline Reload** | 0 | 0.0 KB | 109 | 119 | 129 | 141 | - |
| **Update N->N+1 (JS)** | 86 | 664.3 KB | - | - | - | 8,135 | - |
| **Update N->N+1 (CSS)** | 87 | 664.4 KB | - | - | - | 8,269 | - |
*\*Note: WebKit does not expose `PerformanceLongTaskTiming` (`longtask` entries), so long task total evaluates to 0 ms.*
---
## 4. Variance & Noise Observed
Over the 5 runs per configuration, measurements demonstrated high stability:
- **Wire Bytes & Request Counts**: Deterministic (0% variance across identical runs).
- **Cold Boot-Done**:
- Chromium LTE: Min 10,129 ms, Max 10,798 ms (<6% variance).
- Chromium Lossy: Min 15,871 ms, Max 16,031 ms (<1% variance).
- WebKit LTE: Min 3,138 ms, Max 3,245 ms (<3% variance).
- WebKit Lossy: Min 6,963 ms, Max 8,710 ms (~20% variance due to proxy random stalls).
- **Warm & Offline Reload**:
- Chromium Offline Reload: Min 238 ms, Max 257 ms (<8% variance).
- WebKit Offline Reload: Min 131 ms, Max 141 ms (<7% variance).
- **Update Duration**:
- Chromium: Runs cluster around ~2,000 ms (2,013 ms median). Occasional outlier runs (~45–200 ms) occur if the browser resolves `reg.update()` immediately before network completion reporting.
- WebKit: Steady scaling between LTE (~4.5 s) and Lossy (~8.2 s) pacing.
---
## 5. Limitations & Caveats
1. **WebKit Network Throttling**:
WebKit lacks Chrome DevTools Protocol (CDP) support for hardware network emulation and CPU throttling. We paced throughput (1.6 Mbps) and injected latency/stalls using a custom local reverse proxy (`perf/proxy.mjs`). This throttles HTTP response chunks truthfully but cannot simulate OS-level socket buffer delays or 4× CPU throttling.
2. **Autoplay Policy in WebKit**:
Playwright WebKit on Linux rejects `--autoplay-policy=no-user-gesture-required` and enforces Safari-style gesture requirements on unmuted audio elements. The media playback benchmark detects this and correctly reports `autoplayBlocked: true` without failing the test run.
3. **WebKit SetOffline Incompatibility**:
Calling Playwright's `context.setOffline(true)` on Linux WebKit causes an unrecoverable engine crash (`WebKit encountered an internal error`). Offline mode for WebKit is simulated by terminating the proxy upstream connection (`proxy.setOffline(true)`).
---
## 6. Manual iPhone Testing Protocol (Phase 0 / iOS Safari PWA)
Because automated headless WebKit cannot fully replicate iOS Safari's mobile background constraints and battery telemetry, the following manual protocol is required:
### Protocol Steps
1. **Install PWA**:
- Open Safari on iPhone and navigate to `https://worship.hesed.sbs`.
- Tap the Share button -> **"Add to Home Screen"**.
- Launch ytplayer directly from the Home Screen icon.
2. **Verify Version & Cache**:
- Verify the footer / version string matches `1175f1a1d2c1` (or current production build tag).
- Turn on Airplane Mode and force-reload to confirm offline readiness. Turn Airplane Mode back off.
3. **30-Minute Screen-Off Battery Test**:
- Note baseline battery percentage (e.g., 85%).
- Select a pre-cached playlist or track and tap **Play**. Ensure continuous audio is heard.
- Lock the screen (display off) and let playback run uninterrupted for **30 minutes**.
- Note battery percentage at 30 minutes (calculate drain %/hr).
- Listen for any audio dropouts, stuttering, or premature process suspension.
4. **Interruption & Resumption Test**:
- Trigger an interruption while locked (e.g., initiate an incoming phone call or start Siri).
- End the interruption and observe if playback automatically resumes via `MediaSession` handlers.
- Test Lock Screen Controls (Play/Pause, Next Track, scrubber).
5. **Log Findings**:
- Record iPhone Model, iOS Version, Battery Drain (% per 30m), Dropouts Count (0 = pass), and Resumption Status (Pass/Fail).