Simo/scripts/sync-youtube-videos.mjs

import fs from "node:fs/promises";
import path from "node:path";
import { spawn } from "node:child_process";

const CHANNEL_VIDEOS_URL = "https://www.youtube.com/@simonlol/videos";
const OUT_DIR = path.join(process.cwd(), "src", "content", "blogs");

function run(cmd, args) {
	return new Promise((resolve, reject) => {
		const child = spawn(cmd, args, { stdio: ["ignore", "pipe", "pipe"] });
		let out = "";
		let err = "";
		child.stdout.on("data", (d) => (out += d.toString("utf8")));
		child.stderr.on("data", (d) => (err += d.toString("utf8")));
		child.on("error", reject);
		child.on("close", (code) => {
			if (code === 0) resolve({ out, err });
			else reject(new Error(`${cmd} exited with ${code}\n${err}`));
		});
	});
}

function escYaml(s) {
	return String(s).replaceAll('"', "\\\"");
}

function videoMd({ id, title, uploadDate, viewCount }) {
	const safeTitle = title?.trim() || id;
	// yt-dlp gives upload_date as YYYYMMDD.
	const pubDate = uploadDate && /^\d{8}$/.test(uploadDate)
		? `${uploadDate.slice(0, 4)}-${uploadDate.slice(4, 6)}-${uploadDate.slice(6, 8)}`
		: new Date().toISOString().slice(0, 10);
	return `---\n` +
		`title: "${escYaml(safeTitle)}"\n` +
		`pubDate: ${pubDate}\n` +
		`views: ${Number.isFinite(Number(viewCount)) ? Number(viewCount) : 0}\n` +
		`description: "YouTube video"\n` +
		`---\n\n` +
		`<div style="position:relative; width:100%; aspect-ratio:16/9; background:#000;">\n` +
		`  <iframe\n` +
		`    title="${safeTitle.replaceAll("&", "&amp;").replaceAll("<", "&lt;").replaceAll(">", "&gt;")}"\n` +
		`    src="https://www.youtube-nocookie.com/embed/${id}?rel=0"\n` +
		`    style="position:absolute; inset:0; width:100%; height:100%; border:0;"\n` +
		`    allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"\n` +
		`    allowfullscreen\n` +
		`    referrerpolicy="strict-origin-when-cross-origin"\n` +
		`  ></iframe>\n` +
		`</div>\n`;
}

async function main() {
	// Ensure output dir exists
	await fs.mkdir(OUT_DIR, { recursive: true });

	// Fetch list via yt-dlp.
	// Note: `--flat-playlist` does not include `upload_date` (it returns NA).
	// We fetch ids+titles flat, then fetch upload_date per video id.
	const { out } = await run("yt-dlp", [
		"--flat-playlist",
		"--print",
		"%(id)s\t%(title)s",
		CHANNEL_VIDEOS_URL,
	]);

	const lines = out
		.split("\n")
		.map((l) => l.trim())
		.filter(Boolean);

	const videos = lines
		.map((l) => {
			const [id, ...rest] = l.split("\t");
			const title = rest.join("\t");
			return { id, title };
		})
		.filter((v) => v.id && v.id !== "NA");

	for (const v of videos) {
		try {
			const { out: metaOut } = await run("yt-dlp", [
				"--no-playlist",
				"--print",
				"%(upload_date)s\t%(view_count)s",
				`https://www.youtube.com/watch?v=${v.id}`,
			]);
			const first = metaOut.trim().split("\n")[0] || "";
			const [uploadDate = "", viewCount = ""] = first.split("\t");
			v.uploadDate = uploadDate;
			v.viewCount = viewCount;
		} catch {
			v.uploadDate = "";
			v.viewCount = "";
		}
	}

	// Read existing generated files
	const existing = await fs.readdir(OUT_DIR);
	const existingGenerated = new Set(existing.filter((f) => f.startsWith("video-") && f.endsWith(".md")));

	// Write/update current
	for (const v of videos) {
		const fileName = `video-${v.id}.md`;
		existingGenerated.delete(fileName);
		await fs.writeFile(path.join(OUT_DIR, fileName), videoMd(v), "utf8");
	}

	// Remove stale generated files
	for (const stale of existingGenerated) {
		await fs.unlink(path.join(OUT_DIR, stale));
	}
}

main().catch((e) => {
	console.error("[sync-youtube-videos]", e?.message || e);
	process.exit(1);
});