UCLA Daily: SEIS Fetch

UCLASEISFetch pulls events from UCLA's School of Education & Information Studies - SEIS to the people who live there - by scraping their public events page. If your lobby display or daily brief cares about education talks, pedagogy workshops, or the School's public lectures, this is the node. If it doesn't, skip it; the pack has plenty of other fetchers, and this one is deliberately narrow.

It belongs to the scraper family of the UCLA Daily pack (alongside CS, Fowler, and Hammer), which fetches server-rendered HTML rather than RSS. The pack, by Jeffrey A. Brick, aggregates UCLA campus events and news into 1920x1080 lobby-display cards. Fetcher contract as ever: UCLA_DATA in, UCLA_DATA plus segment_text out, chainable in any order between UCLADailyDataInit and the card, narrator, or pulse nodes.

How it works

It fetches https://seis.ucla.edu/events/ by default and extracts event entries from the HTML. The node uses the shared http_utils helper - requests with a urllib fallback - so it runs in a bare environment, and a failed scrape is logged to errors[] rather than crashing the pipeline. IS_CHANGED means it re-fetches on every queue run, keeping the display current at the cost of being chatty on the network.

The usual scraper caveats apply and are worth internalizing: this node's correctness depends entirely on the SEIS site's markup staying the same. When a university department redesigns its events page - and they all do, eventually - the extraction can quietly return nothing. That's the expected failure mode, not a bug in your install.

The inputs

• ucla_data - required pipeline dict from UCLADailyDataInit.
• url - defaults to the SEIS events page; override if the school moves.
• max_items - 1–30, default 10.

Outputs

Two: ucla_data (enriched, chainable) and segment_text - the plain-text event list, ready for UCLALobbyCard or UCLANarrator.

Install

ComfyUI Manager → search UCLA Daily, or:

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

Restart ComfyUI. No API keys, no model downloads.

Common issues

The pattern is the same across the scraper family: watch for site redesigns, and don't re-run constantly while you tweak card layout - every queue run is a live scrape of a third-party site. Academic calendars also go quiet over breaks; an empty result in December is probably a real empty calendar. If you want coverage that's more robust to markup changes, the RSS-based fetchers (Newsroom, Med, Arts) are the sturdier siblings.