AirwaveDB imports open directory data and keeps upstream identifiers in data/source_map.json.
- Website: https://www.radio-browser.info/
- API documentation: https://docs.radio-browser.info/
- Role: primary station discovery and metadata source
- Imported fields: station UUID, name, country, subdivision, languages, tags, homepage, favicon, popularity metadata, and update timestamps
AirwaveDB queries entries reported as working by the directory (hidebroken=true) and uses a descriptive User-Agent. Multiple Radio Browser mirrors are tried before a sync is considered unavailable.
- Repository: https://github.com/eyalev/verified-radio-streams
- Role: independently verified stream source and secondary metadata source
- License: CC0 1.0
The dataset contains stream endpoints that returned audio during its verification run. AirwaveDB preserves the upstream verification date and source ID.
No upstream source is treated as authoritative for every field. AirwaveDB groups matching source records into a stable station entity, preserves provenance, and selects display metadata deterministically. Existing source-to-station mappings are reused so normal upstream metadata changes do not unnecessarily change AirwaveDB station IDs.
A scheduled rebuild is rejected when a required upstream source is unavailable or when the resulting dataset shrinks unexpectedly. This keeps a temporary upstream failure from replacing a healthy snapshot.
Missing metadata is left empty rather than guessed. Station and stream records can change independently: stream URLs are stored in AirwaveDB/streams, not in this repository.