{"slug": "how-to-export-any-public-telegram-channel-to-json-and-read-it-in-english-no-no", "title": "How to export any public Telegram channel to JSON and read it in English (no login, no bot token)", "summary": "A developer built a Python tool that exports any public Telegram channel to JSON without a login or bot token by scraping the t.me/s/ web preview with httpx and BeautifulSoup, then paginating backwards via the ?before= parameter. The writeup documents five ways real channels broke the initial parser — reply blocks containing nested message text, video and sticker elements, abbreviated view counts like \"15.4K\", and HTTP 302 redirects for channels without a public preview — and describes adding LLM-based translation and per-post summaries batched into a single JSON response to cut costs.", "body_md": "A lot of useful information lives in Telegram channels: tech news, market commentary, local announcements. Much of it isn't in English. If you want to follow ten Russian or Ukrainian tech channels, copying posts into a translator one by one gets old fast.\n\nThe obvious options both have catches:\n\n`api_id` from my.telegram.org and a login with your phone number. That's more setup than a quick export deserves, and it ties the scraper to your personal account.\nThere's a third option that needs neither: every public channel has a web preview at `https://t.me/s/`. It's plain HTML, about 20 posts per page, and it pages backwards with a `?before=` parameter.\n\nI built a small tool on top of it. Here is how it works, the problems I ran into, and how I added AI translation and summaries.\n\n``` python\nimport re\nimport time\n\nimport httpx\nfrom bs4 import BeautifulSoup\n\ndef parse_page(html: str):\n soup = BeautifulSoup(html, \"html.parser\")\n posts = []\n for msg in soup.select(\".tgme_widget_message[data-post]\"):\n if \"service_message\" in msg.get(\"class\", []):\n continue\n text_el = msg.select_one(\".tgme_widget_message_text.js-message_text\")\n if text_el:\n for br in text_el.find_all(\"br\"):\n br.replace_with(\"\\n\")\n time_el = msg.select_one(\".tgme_widget_message_date time\")\n views_el = msg.select_one(\".tgme_widget_message_views\")\n posts.append({\n \"id\": int(msg[\"data-post\"].split(\"/\")[1]),\n \"url\": \"https://t.me/\" + msg[\"data-post\"],\n \"date\": time_el[\"datetime\"] if time_el else None,\n \"text\": text_el.get_text().strip() if text_el else \"\",\n \"views\": views_el.get_text(strip=True) if views_el else None, # e.g. \"15.4K\"\n })\n more = soup.select_one(\"link[rel=prev]\") # href=\"https://dev.to/s/?before=\"\n before = re.search(r\"before=(\\d+)\", more[\"href\"]).group(1) if more else None\n return posts, before\n\ndef fetch_channel(channel: str, limit: int = 100):\n url, posts = f\"https://t.me/s/{channel}\", []\n with httpx.Client(headers={\"User-Agent\": \"Mozilla/5.0\"}, follow_redirects=False) as client:\n while url and len(posts) < limit:\n resp = client.get(url)\n if resp.status_code != 200: # 302 = no public preview for this channel\n break\n page, before = parse_page(resp.text)\n posts += sorted(page, key=lambda p: p[\"id\"], reverse=True)\n url = f\"https://t.me/s/{channel}?before={before}\" if before else None\n time.sleep(1) # be polite\n return posts[:limit]\n\nif __name__ == \"__main__\":\n for post in fetch_channel(\"durov\", limit=5):\n print(post[\"date\"], post[\"views\"], post[\"text\"][:80])\n```\n\n`pip install httpx beautifulsoup4` and it runs. Each page lists posts oldest-first, so I sort each page newest-first and then follow `before` to older pages.\n\nThe parser above worked on my first test page. Real channels then broke it in five ways.\n\n`.tgme_widget_message_reply` block, and that block also contains a `.tgme_widget_message_text`. Grab the first match and you save the `.js-message_text`, or skip anything inside the reply block.`