generated: '2026-09-07'
method: derived
source: >-
  Derived from components.schemas and the path structure of
  openapi/eu-open-data-portal-hub-search-openapi.yaml and openapi/eu-open-data-portal-hub-repo-openapi.yaml,
  read against the DCAT-AP application profile the contracts declare.
provider: eu-open-data-portal
model: DCAT-AP
model_note: >-
  The data model is not bespoke: it is DCAT-AP, the European application profile of W3C DCAT, and the JSON
  the search API returns is a flattened projection of the RDF graph the registry stores. Every reference
  field below is a URI into the same graph (http://data.europa.eu/88u/...), which is why the registry can
  serve any resource as RDF/XML, Turtle, JSON-LD, N-Triples, N-Quads, TriG, TriX or N3.
identifier_scheme:
  form: 'http://data.europa.eu/88u/<type>/<id>'
  examples:
  - http://data.europa.eu/88u/catalogue/eurogeographics
  - http://data.europa.eu/88u/dataset/dataset-status~~1
  note: >-
    Within a catalogue a publisher additionally chooses an `originalId`; the pair (catalogueId, originalId)
    is the upsert key on the /catalogues/{catalogueId}/resources/origin path.
entities:
- name: Catalogue
  description: A federated data catalogue — a national portal, an EU institution or a thematic collection.
  schema: components.schemas.Catalogue (hub-search)
  key_fields: [id, title, description, publisher, country, homepage, license, theme_taxonomy, source_type, catalogueProfile]
  relationships:
  - has_many: Dataset
    via: has_part / dcat:dataset
  - has_many: DCATResource
    via: /catalogues/{catalogueId}/resources
  - belongs_to: Organization
    via: publisher
  - belongs_to: Catalogue
    via: is_part_of
    note: catalogues nest — a super-catalogue can contain member catalogues
- name: Dataset
  description: A DCAT-AP dataset record, the unit of the ~1.9M-record corpus.
  schema: components.schemas.Dataset (hub-search)
  key_fields: [id, title, description, keywords, publisher, catalog, country, spatial, temporal, issued, modified, language, is_hvd, hvd_category, applicable_legislation, quality_meas]
  relationships:
  - has_many: Distribution
    via: distributions
  - belongs_to: Catalogue
    via: catalog
  - belongs_to: Organization
    via: publisher
  - has_one: CatalogRecord
    via: catalog_record
  - has_many: Vocable
    via: 'subject / theme / hvd_category / country / language (every one is a URI into a controlled vocabulary)'
  - has_many: DatasetRevision
    via: /datasets/{id}/revisions/{revisionId}
  - has_one: QualityMeasurement
    via: quality_meas
  - belongs_to: DatasetSeries
    via: is_version_of / has_version
- name: Distribution
  description: A downloadable or callable representation of a dataset.
  schema: components.schemas.Distribution (hub-search)
  key_fields: [id, title, access_url, download_url, format, media_type, byte_size, checksum, license, status, availability, conforms_to]
  relationships:
  - belongs_to: Dataset
    via: 'parent path /datasets/{datasetId}/distributions'
  - has_one: DataService
    via: access_service
  - has_one: QualityMeasurement
    via: /distributions/{id} on the MQA metrics cache
- name: DataService
  description: A DCAT data service — an API endpoint catalogued as a first-class resource.
  schema: components.schemas.DataService (hub-search)
  key_fields: [id, title, endpoint_url, endpoint_description, serves_dataset, conforms_to, access_right, license]
  relationships:
  - has_many: Dataset
    via: serves_dataset
  - belongs_to: Catalogue
    via: catalog
- name: DatasetSeries
  description: A series that groups successive dataset versions.
  schema: 'hub-search /datasetseries/{id}; hub-repo DCAT Resources'
  relationships:
  - has_many: Dataset
    via: 'num_series / is_version_of'
- name: Organization
  description: A publishing organisation (corporate body), with a resolvable ancestry hierarchy.
  schema: 'hub-search /organizations, /organizations?hierarchy=true'
  relationships:
  - has_many: Catalogue
    via: publisher
  - belongs_to: Organization
    via: 'ancestors (added in hub-search 5.3.11)'
  - has_one: Vocable
    via: corporate-body-classification
- name: Vocabulary
  description: A controlled vocabulary (authority table) — 26 are served, EuroVoc the largest at 7,486 concepts.
  schema: components.schemas.Vocabulary (hub-search)
  relationships:
  - has_many: Vocable
    via: /vocabularies/{vocabulary}/{id}
  cross_reference: vocabulary/eu-open-data-portal-vocabularies.yml
- name: Vocable
  description: One concept inside a vocabulary, carrying multilingual pref_labels and an optional parent_id.
  schema: components.schemas.Vocable (hub-search)
  relationships:
  - belongs_to: Vocabulary
    via: 'path parameter {vocabulary}'
  - belongs_to: Vocable
    via: parent_id
- name: QualityMeasurement
  description: >-
    An MQA measurement over a catalogue, country, dataset or distribution, scored across findability,
    accessibility, interoperability, reusability and contextuality out of 450 points.
  schema: 'MQA metrics cache /global, /countries/{id}, /catalogues/{id}, /datasets/{id}, /distributions/{id}'
  relationships:
  - belongs_to: [Catalogue, Dataset, Distribution, Country]
    via: 'scope + id in the metrics path'
  - has_many: Violation
    via: /catalogues/{id}/violations
- name: DatasetRevision
  description: A stored prior state of a dataset record; readable, diffable and individually deletable.
  schema: 'hub-search /datasets/{id}/revisions/{revisionId}; hub-repo /datasets/revisions/{revisionId}, /datasets/{datasetId}/diff'
  relationships:
  - belongs_to: Dataset
    via: '{id}'
- name: EditorialContent
  description: Portal editorial pages (news, data stories) indexed alongside data records.
  schema: components.schemas.EditorialContent (hub-search)
- name: Draft
  description: An unpublished dataset held in the registry until published or hidden.
  schema: 'hub-repo /drafts/datasets'
  relationships:
  - becomes: Dataset
    via: 'PUT /drafts/datasets/publish/{id}'
counts:
  entities: 13
  schemas_in_hub_search: 15
render: null