Skip to main content

Library to integrate the MoJ data platform with the catalogue component.

Project description

Data platform catalogue

This library is part of the Ministry of Justice data platform.

It publishes object metadata to a data catalogue, so that the metadata can be made discoverable by consumers.

Broadly speaking, a catalogue stores a metadata graph, consisting of data assets. Data assets could be tables, schemas or databases.

How to install

To install the package using pip, run:

pip install ministryofjustice-data-platform-catalogue

Terminology

  • Data assets - Any databases, tables, or schemas within the metadata graph
  • Data products - Groupings of data assets that are published for reuse across MOJ. In the data platform, the concepts of database and data product are similar, but they may be represented as different entities in the catalogue.
  • Domains - allow metadata to be grouped into different service areas that have their own governance, like HMCTS, HMPPS, OPG, etc.

Example usage

from data_platform_catalogue import (
  DataHubCatalogueClient,
  BaseCatalogueClient, DataLocation, CatalogueMetadata,
  DataProductMetadata, TableMetadata,
  CatalogueError
)

client: BaseCatalogueClient = DataHubCatalogueClient(jwt_token=jwt_token, api_url=api_url)

data_product = DataProductMetadata(
    name = "my_data_product",
    description = "bla bla",
    version = "v1.0.0",
    owner = "7804c127-d677-4900-82f9-83517e51bb94",
    email = "justice@justice.gov.uk",
    retention_period_in_days = 365,
    domain = "HMCTS",
    dpia_required = False
)

table = TableMetadata(
  name = "my_table",
  description = "bla bla",
  column_details=[
      {"name": "foo", "type": "string", "description": "a"},
      {"name": "bar", "type": "int", "description": "b"},
  ],
  retention_period_in_days = 365,
  major_version = 1
)

try:
    table_fqn = client.upsert_table(
        metadata=table,
        data_product_metadata=data_product,
        location=DataLocation("test_data_product_v1"),
    )
except CatalogueError:
  print("oh no")

Catalogue Implementations

DataHub

  • Each data product within the MOJ data platform is created as a data product entity
  • Each table is created as a dataset in DataHub
  • Tables that reside in the same athena database (data_product_v1) should be placed within the same DataHub container.

OpenMetadata

  • Each MOJ data product is mapped to a database in the OpenMetadata catalogue
  • We populate the schema level in openmetdata with a generic entry of Tables
  • Each table is mapped to a table in openmetadata

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

Built Distribution

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page