orbit/dev/wdshot.mjs

145 lines
5.5 KiB
JavaScript

// One-shot headless-Firefox screenshot via geckodriver (W3C WebDriver).
//
// node dev/wdshot.mjs <url> <out.png> [setupExpression] [timeoutMs=20000] [settleMs=0]
//
// The setup expression (optional) runs right before the capture — it may
// return a Promise; the runner waits for it to settle (so scene state /
// animations can be driven first). WebDriver's /screenshot endpoint is
// used (not canvas.toDataURL — a WebGL canvas without
// preserveDrawingBuffer captures black).
//
// settleMs (optional): a runner-side settle wait AFTER navigation, before
// the setup script runs. Pages with async module boot (top-level awaits —
// e.g. dev/popup-shot.mjs waiting for the web fonts) need a beat before
// their globals exist.
import { spawn, execFileSync } from 'node:child_process';
import { existsSync, readdirSync, writeFileSync } from 'node:fs';
const [url, out, setup, timeoutMs = '20000', settleMs = '0'] = process.argv.slice(2);
if (!url || !out) {
console.error('usage: node dev/wdshot.mjs <url> <out.png> [setupExpression] [timeoutMs] [settleMs]');
process.exit(2);
}
const sleep = (ms) => new Promise((r) => setTimeout(r, ms));
let geckodriver;
try {
geckodriver = execFileSync('sh', ['-c', 'command -v geckodriver'], { encoding: 'utf8' }).trim();
} catch {
console.error('need geckodriver on PATH');
process.exit(2);
}
const firefoxCandidates = [];
try {
const ff = execFileSync('sh', ['-c', 'command -v firefox'], { encoding: 'utf8' }).trim();
if (ff) firefoxCandidates.push(ff);
} catch { /* not on PATH */ }
try {
const snapRev = readdirSync('/snap/firefox', { withFileTypes: true }).map((d) => d.name);
for (const rev of [null, ...snapRev]) {
const base = rev ? `/snap/firefox/${rev}` : '/snap/firefox/current';
const p = `${base}/usr/lib/firefox/firefox`;
if (existsSync(p)) firefoxCandidates.push(p);
}
} catch { /* non-snap install */ }
if (firefoxCandidates.length === 0) {
console.error('no firefox binary found');
process.exit(2);
}
const PORT = 4444 + (process.pid % 100);
const driver = spawn(geckodriver, ['--port', String(PORT), '--log', 'fatal'], { stdio: 'ignore' });
const base = `http://127.0.0.1:${PORT}`;
const fail = (msg, code = 1) => {
try { driver.kill('SIGKILL'); } catch { /* dead */ }
console.error(msg);
process.exit(code);
};
const wd = async (method, path, body) => {
const res = await fetch(base + path, {
method,
headers: { 'Content-Type': 'application/json' },
body: body === undefined ? undefined : JSON.stringify(body),
});
const text = await res.text();
let data = null;
try { data = JSON.parse(text); } catch { /* empty */ }
if (!res.ok) throw new Error(`WebDriver ${method} ${path}${res.status}: ${text.slice(0, 300)}`);
return data;
};
try {
let up = false;
for (let i = 0; i < 60; i++) {
try {
const v = await (await fetch(`${base}/status`)).json();
if (v?.value?.ready || v?.status === 'ok' || v?.value) { up = true; break; }
} catch { /* not up */ }
await sleep(250);
}
if (!up) fail('geckodriver never came up');
let sess;
let lastErr = null;
for (const ff of firefoxCandidates) {
try {
sess = await wd('POST', '/session', {
capabilities: {
alwaysMatch: {
browserName: 'firefox',
'moz:firefoxOptions': { args: ['-headless'], binary: ff },
// This box is CPU-starved: boot + animation can outlive the
// driver's 30 s default. Raise the script budget.
timeouts: { script: 180000 },
},
},
});
break;
} catch (err) { lastErr = String(err?.message ?? err); }
}
if (!sess) fail(`no firefox candidate accepted: ${lastErr}`);
const sid = sess.value?.sessionId;
if (!sid) fail('no session id');
// Belt & braces: also set the session script timeout explicitly.
await wd('POST', `/session/${sid}/timeouts`, { script: 180000 }).catch((e) => {
console.error('timeouts: ' + String(e?.message ?? e).slice(0, 120));
});
await wd('POST', `/session/${sid}/url`, { url });
// Settle: let the page finish its async boot (module top-level awaits)
// before the setup script runs — a runner-side wait, since the page's
// own timers may be starved in headless.
if (Number(settleMs) > 0) await sleep(Number(settleMs));
if (setup) {
// Multiple scripts, each held as a pending Promise under execute/sync
// (geckodriver honours the session script timeout set above — 3 min).
// setup may be a single script string or a JSON array of them.
let steps = [setup];
try { const parsed = JSON.parse(setup); if (Array.isArray(parsed)) steps = parsed; } catch { /* single script */ }
for (let i = 0; i < steps.length; i++) {
const script = `return (async () => { ${steps[i]} })();`;
try {
const out = await wd('POST', `/session/${sid}/execute/sync`, { script, args: [] });
if (out?.value !== undefined && out.value !== null) {
console.log('setup[' + i + '] → ' + JSON.stringify(out.value).slice(0, 400));
}
} catch (err) {
console.error(`setup[${i}]: ` + String(err?.message ?? err).slice(0, 200));
}
}
}
const shot = await wd('GET', `/session/${sid}/screenshot`);
if (!shot?.value) fail('no screenshot in response');
const b64 = String(shot.value).replace(/^data:image\/\w+;base64,/, '');
writeFileSync(out, Buffer.from(b64, 'base64'));
console.log('saved ' + out);
await wd('DELETE', `/session/${sid}`).catch(() => {});
} finally {
try { driver.kill('SIGKILL'); } catch { /* dead */ }
}