Read metadata/distribution.csv, the curated endpoints per dataset
Source: R/catalog_datasets.R
read_distribution_registry.RdOne row per external endpoint a dataset can be got or seen through that the
release cannot measure itself — a CoastWatch mirror, the EDI or NCEI record
the ingest read from, the OBIS dataset and its IPT resource, a legacy ERDDAP
id with its sunset — with kind, portal, id, url, title, status
(current | superseded | retired | external | planned), superseded_by (a
URL or a dataset_key), observed_utc (when the status was last confirmed;
observe_distributions() will refresh it) and notes. Nothing is ever
deleted from it: a dead endpoint becomes retired with the date, so a page
can say "was at X until …". Every vocabulary column is validated; an unknown
value errors.
Value
A tibble, all columns character.