Skip to main content
The data catalog is the inventory of logical tables available to a Workspace. It brings four pools into one SQL connection while preserving the owner and lifecycle of each table. Provider quality lives beside the pools at providers_quality. It measures Provider tables rather than holding business records.

​
What the catalog describes

For each logical table, the catalog can describe:
  • Its catalog, schema, table name, and table type.
  • Its Arrow schema and semantic scalar annotations.
  • Primary and foreign-key relationships.
  • The Provider stream, artifact, or Plot that produces it.
  • Shape, freshness, quality, and lineage when those annotations are available.
An omitted annotation means that the value is not known or has not been computed. A present value is authoritative.
A table can be registered before it contains rows. Discovering a table proves that it is addressable; it does not prove that a sync, publication, or Plot run has produced data.

​
Discover before querying

Use standard Flight SQL metadata commands for portable catalog, schema, table, column, and key discovery. Use the Workspace API’s catalogV2.catalog operation when you also need Parable-specific ownership and annotations. Discover tables and columns shows both paths.

​
Protected Workspace statistics

Catalog table statistics require a valid storage key lease for every protected bucket in the Workspace catalog. This check runs before reading cached row counts, byte sizes, or quality status. A cached answer does not bypass a refused or expired key lease. Ordinary Workspaces need no lease.

Discover Tables and Columns

Inspect Workspace catalogs, schemas, tables, columns, and semantic annotations before writing SQL.

Search Available Data

Find the tables, columns, and fields that hold the data a question needs with search_available_data, a search of the Workspace and Provider Pools by keyword and, where it is configured, by meaning.

Search Everything

Find Provider records by what their text says with search_everything, and read the matching rows from their source tables.

Provider Pool

Query and join the current data supplied by Provider Plugins.

Artifact Pool

Query files that a Workspace has published as SQL tables, and the data files of each Parable.

Parable Pool

Query materialized Plot results by Parable and Plot slug.

Workspace Pool

Query live Workspace membership, Provider, role, Parable, Plot, and run metadata.

Provider Quality

Query checks, statistical process control, shape, and health for Provider tables.