Cleaning Cinema Titles Before You Can Even Search

Clusterflick scrapes 250+ cinema websites but must first normalise messy film titles before TMDB matching. A pipeline strips venue branding, event labels, and separators using ~1,000 known phrases and correction lists, backed by nearly 15,000 test cases. Edge cases like legitimate '+' in titles (e.g. Romeo + Juliet) require special handling to avoid false splits.