UCLA Daily: Film Archive Fetch

Here's the thing you need to know first: the UCLA Film & Television Archive's own events page is a JavaScript-rendered single-page app. You cannot scrape it with a regex and a prayer, and this node doesn't try. UCLAFilmArchiveFetch does the honest workaround - it pulls the main UCLA Newsroom RSS feed and filters items for film and archive keywords. If the Archive screens a restored film, the Newsroom publishes a story about it, and that story is what lands on your lobby screen.

This is the "we couldn't scrape the real source, so we found the next best public signal" node of the UCLA Daily pack, and honestly it's a good lesson in how the pack thinks. Every fetcher takes UCLA_DATA in and hands it back enriched, plus a segment_text string, so you can chain any mix of fetchers between UCLADailyDataInit and the card or pulse nodes. Film Fetch slots in when you want the cinema-curious residents to see what's playing or being preserved.

How it works

The default feed is the Newsroom main RSS (https://newsroom.ucla.edu/rss.xml). Items are filtered against a keyword list - film, archive, screening, cinema, movie, television, billy wilder, preservation, restoration, festival of preservation, documentary. If nothing matches, the node logs a note and moves on rather than inventing content. That's the pack's graceful-failure style: no crash, no fake data, just an entry in errors[] and empty hands.

As with the other fetchers, HTTP goes through the shared helper that tries requests and falls back to urllib, and IS_CHANGED forces a fresh pull every queue run. If you want the Archive's full event listings rather than news-adjacent stories, you'd need a real scraper - the author explicitly says the Nuxt/GraphQL page is out of reach for a stdlib node, so this is the pragmatic compromise, not a bug.

The inputs

• ucla_data - the pipeline dict from UCLADailyDataInit. Required, as with all fetchers.
• rss_url - default https://newsroom.ucla.edu/rss.xml. Point it at another Newsroom feed if you want to widen or narrow the haystack.
• max_items - 1–20, default 5. Cap on film-related items included.

Outputs

Two: ucla_data (chain onward or into a card) and segment_text - plain text of the film items, ready for UCLALobbyCard or the UCLANarrator to rewrite.

Install

ComfyUI Manager, search UCLA Daily, or by hand:

cd ComfyUI/custom_nodes
git clone https://github.com/jbrick2070/ComfyUI-UCLADaily.git
pip install qrcode pillow requests

Restart ComfyUI. No model downloads, no API keys.

Common issues

Don't expect this node to be exhaustive. It surfaces film news when the Newsroom publishes it - which, for a busy Archive season, is a lot, but on a slow week it may come back empty. That's not a broken install; that's the nature of filtering someone else's feed. If you need guaranteed Archive listings, the pack has no node for that yet - this is the best public-data approximation the author shipped.