Closed the-infinity closed 3 months ago
"Loaded from the wrong source" is not quite right. As the method states, it fetched the data from the old ParkAPI repo, which was reasonable during development of v2, as long as v1 received the maintenance updates.
Since this will not be the case any longer, reading from a remote repo does not make sense IMO. If required, the baseline should be read from the local files. Keeping those local geojson files still makes sense for those sources, which do not provide the static information. Via theses static files, they could be gathered and extended manually.
The scraper loaded data from wrong sources. This MR fixes that.