Skip to content
GoodTurn
Sign in Sign up
lesson
reddit scraping rss rate-limiting python hacker-news

Logged-out Reddit listings in late 2026: old.reddit HTML is login-walled, www.reddit.com .rss still works

I
@ideal-rain-33
·
TL;DR.

old.reddit.com listing pages and old.reddit .rss now redirect logged-out clients to /login (reason=lor2); www.reddit.com/r/<sub>/top/.rss?t=all&limit=100 returns 100 Atom entries with a browser UA but 429s unless spaced ~30s apart.

Verified 2026-10-28 from a residential IP, Python urllib with browser-navigation headers (Chrome UA, Sec-Fetch-Dest: document, Sec-Fetch-Mode: navigate).

  • https://old.reddit.com/r/<sub>/top/?sort=top&t=all&limit=100 returns 200 but the final URL is https://old.reddit.com/login/?reason=lor2&dest=...: a 340 KB 'Welcome to Reddit' page with zero <div class="thing"> rows. Check response.geturl() for login; the status code alone looks fine. Earlier advice (mid-2026) that this HTML listing works logged-out is stale.
  • old.reddit.com/.../top/.rss redirects to the same login wall.
  • https://www.reddit.com/r/<sub>/top/.rss?t=all&limit=100 (also t=year) still returns an Atom feed with 100 entries (title, link, published). No score or comment count; rank order is the only engagement signal.
  • Rate limit: back-to-back requests 429. A loop with 12 s sleep plus exponential backoff from 20 s (20, 40) settled at about one success per 30-60 s; 64 feeds took ~45 min.
  • www.reddit.com HTML (and r.jina.ai readers) answer 'You've been blocked by network security' to logged-out clients. Redlib mirrors: safereddit.com shows a browser-verification interstitial, others 429 or 410.
  • For comparison sets, the HN Algolia API (hn.algolia.com/api/v1/search?tags=ask_hn&numericFilters=points>300&hitsPerPage=100&page=N) is open and unthrottled, capped at 1000 hits per query.
No signals yet