aishare/scripts/debug-chatgpt-share.mjs

import { chromium } from "playwright";
import { spawn } from "node:child_process";
import { existsSync, readdirSync } from "node:fs";
import { mkdtemp, rm } from "node:fs/promises";
import net from "node:net";
import os from "node:os";
import path from "node:path";

const sourceUrl = process.argv[2];
const holdMs = Number(process.env.DEBUG_HOLD_MS || "900000");
const port = Number(process.env.DEBUG_PORT || "9222");
const proxyPort = process.env.DEBUG_PROXY_PORT ? Number(process.env.DEBUG_PROXY_PORT) : null;
const cdpOnly = process.env.DEBUG_CDP_ONLY === "1";

if (!sourceUrl) {
  console.error("usage: node scripts/debug-chatgpt-share.mjs <chatgpt-share-url>");
  process.exit(2);
}

let browser;
let context;
let browserProcess;
let proxyServer;
let userDataDir;

function now() {
  return new Date().toISOString();
}

function findChromiumExecutable() {
  const preferred = process.env.DEBUG_CHROMIUM_EXECUTABLE || chromium.executablePath();
  if (existsSync(preferred)) return preferred;

  const browsersRoot = process.env.PLAYWRIGHT_BROWSERS_PATH || path.dirname(path.dirname(preferred));
  const headlessShell = readdirSync(browsersRoot)
    .filter((entry) => entry.startsWith("chromium_headless_shell-"))
    .sort()
    .reverse()
    .map((entry) => path.join(browsersRoot, entry, "chrome-headless-shell-linux64", "chrome-headless-shell"))
    .find((candidate) => existsSync(candidate));

  if (headlessShell) return headlessShell;
  throw new Error(`No Chromium executable found. Tried ${preferred} and ${browsersRoot}/chromium_headless_shell-*.`);
}

async function close() {
  if (context) await context.close().catch(() => undefined);
  if (browser) await browser.close().catch(() => undefined);
  if (proxyServer) await new Promise((resolve) => proxyServer.close(resolve)).catch(() => undefined);
  if (browserProcess && !browserProcess.killed) browserProcess.kill("SIGTERM");
  if (userDataDir) await rm(userDataDir, { recursive: true, force: true }).catch(() => undefined);
}

process.on("SIGINT", () => close().finally(() => process.exit(130)));
process.on("SIGTERM", () => close().finally(() => process.exit(143)));

userDataDir = await mkdtemp(path.join(os.tmpdir(), "aishare-debug-"));
const executablePath = findChromiumExecutable();
console.log(`[${now()}] Chromium executable: ${executablePath}`);
browserProcess = spawn(executablePath, [
  "--headless",
  "--no-sandbox",
  "--disable-dev-shm-usage",
  "--disable-gpu",
  "--hide-scrollbars",
  "--mute-audio",
  "--window-size=1360,900",
  "--remote-debugging-address=0.0.0.0",
  `--remote-debugging-port=${port}`,
  `--user-data-dir=${userDataDir}`,
  cdpOnly ? sourceUrl : "about:blank"
], {
  stdio: ["ignore", "pipe", "pipe"]
});

browserProcess.stdout.on("data", (data) => process.stdout.write(data));
browserProcess.stderr.on("data", (data) => process.stderr.write(data));
browserProcess.on("exit", (code, signal) => {
  console.log(`[${now()}] Chromium exited code=${code} signal=${signal}`);
});

let devtoolsReady = false;
for (let attempt = 0; attempt < 100; attempt += 1) {
  const response = await fetch(`http://127.0.0.1:${port}/json/version`).catch(() => null);
  if (response?.ok) {
    devtoolsReady = true;
    break;
  }
  await new Promise((resolve) => setTimeout(resolve, 250));
}
if (!devtoolsReady) throw new Error(`Chromium DevTools endpoint did not start on 127.0.0.1:${port}.`);

if (proxyPort) {
  proxyServer = net.createServer((client) => {
    const upstream = net.connect(port, "127.0.0.1");
    client.pipe(upstream);
    upstream.pipe(client);
    client.on("error", () => upstream.destroy());
    upstream.on("error", () => client.destroy());
  });
  await new Promise((resolve, reject) => {
    proxyServer.once("error", reject);
    proxyServer.listen(proxyPort, "0.0.0.0", resolve);
  });
  console.log(`[${now()}] DevTools proxy: 0.0.0.0:${proxyPort} -> 127.0.0.1:${port}`);
}

if (cdpOnly) {
  console.log(`[${now()}] CDP-only mode; holding browser for ${Math.round(holdMs / 1000)}s.`);
  await new Promise((resolve) => setTimeout(resolve, holdMs));
  await close();
  process.exit(0);
}

browser = await chromium.connectOverCDP(`http://127.0.0.1:${port}`);

context = await browser.newContext({
  viewport: { width: 1360, height: 900 },
  deviceScaleFactor: 1,
  locale: "en-US"
});

const page = await context.newPage();
page.setDefaultTimeout(30_000);

console.log(`[${now()}] Chromium DevTools port: ${port}`);
console.log(`[${now()}] Opening ${sourceUrl}`);

await page.goto(sourceUrl, { waitUntil: "commit", timeout: 45_000 }).catch((error) => {
  console.log(`[${now()}] goto did not complete: ${error.message}`);
});
await page.waitForLoadState("domcontentloaded", { timeout: 20_000 }).catch(() => undefined);
await page.waitForLoadState("networkidle", { timeout: 10_000 }).catch(() => undefined);
await page.waitForTimeout(3000);

const summary = await page.evaluate(() => {
  const selectors = [
    "[data-message-author-role]",
    "[data-message-id]",
    "[data-testid]",
    "main article",
    "main [role=listitem]",
    "main .markdown",
    "main [class*=markdown]",
    "script[type='application/json']",
    "script#__NEXT_DATA__"
  ];

  const counts = Object.fromEntries(selectors.map((selector) => [selector, document.querySelectorAll(selector).length]));
  const title = document.querySelector("h1")?.textContent?.trim() || document.title;
  const bodySample = (document.body.textContent || "").replace(/\s+/g, " ").trim().slice(0, 600);
  const testIds = Array.from(document.querySelectorAll("[data-testid]"))
    .map((element) => element.getAttribute("data-testid"))
    .filter(Boolean)
    .slice(0, 80);

  return {
    href: location.href,
    title,
    counts,
    testIds,
    bodySample
  };
});

console.log(JSON.stringify(summary, null, 2));
console.log(`[${now()}] Holding browser for ${Math.round(holdMs / 1000)}s.`);

await new Promise((resolve) => setTimeout(resolve, holdMs));
await close();