scientia-website/scripts/fetch-telegram.mjs

// Fetches the public Telegram channel feed (t.me/s/...) and saves the latest
// posts for the home-page Telegram widget. Run manually: npm run fetch-news
import * as cheerio from "cheerio";
import { mkdir, writeFile } from "node:fs/promises";
import path from "node:path";

const CHANNEL = "scientia_ru";
const UA =
  "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/126.0.0.0 Safari/537.36";

const res = await fetch(`https://t.me/s/${CHANNEL}`, { headers: { "User-Agent": UA } });
if (!res.ok) {
  console.error(`Failed to fetch channel: HTTP ${res.status}`);
  process.exit(1);
}
const $ = cheerio.load(await res.text());

const posts = [];
$(".tgme_widget_message").each((_, el) => {
  const $el = $(el);
  const textNode = $el.find(".tgme_widget_message_text").first();
  // Keep basic formatting tags, drop telegram-internal spans/attrs
  textNode.find("a, b, i, u, s, strong, em, code, br").each((__, keep) => {
    const attrs = $(keep).attr() || {};
    for (const name of Object.keys(attrs)) {
      if (name !== "href") $(keep).removeAttr(name);
    }
  });
  const text = textNode.html()?.trim() || "";
  const datetime = $el.find("time").first().attr("datetime") || "";
  const photoStyle = $el.find(".tgme_widget_message_photo_wrap").first().attr("style") || "";
  const photo = photoStyle.match(/url\('([^']+)'\)/)?.[1]?.split("?")[0] || "";
  const link = $el.find(".tgme_widget_message_date").first().attr("href") || `https://t.me/${CHANNEL}`;
  if (text || photo) posts.push({ text, datetime, photo, link });
});

const latest = posts.slice(-4).reverse();
await mkdir(path.dirname("src/data/telegram-news.json"), { recursive: true });
await writeFile("src/data/telegram-news.json", JSON.stringify(latest, null, 2));
console.log(`saved ${latest.length} posts`);

// Channel avatar for the widget header
const avatar = await fetch(`https://t.me/i/userpic/320/${CHANNEL}.jpg`, { headers: { "User-Agent": UA } });
if (avatar.ok) {
  const buf = Buffer.from(await avatar.arrayBuffer());
  if (buf.length > 500) {
    await writeFile("public/images/telegram-avatar.jpg", buf);
    console.log("avatar saved");
  } else {
    console.log("avatar looks empty, skipped");
  }
} else {
  console.log(`avatar fetch failed: HTTP ${avatar.status}`);
}