LakeQL
Overview
  • Introduction
  • Datasource Contract
  • API Reference
Writing
  • Write Strategies
  • Load Strategies
  • Partitioning
Storage
  • Storage Operations
Trino
  • Connection
  • Write Config
  • Metadata
  • API Reference
Fabric
  • Connection
  • Metadata
  • API Reference
GitHub
LakeQL
  1. Data Sources
  2. Fabric
  3. Metadata

Metadata

How the Fabric adapter exposes catalogs, schemas, tables, and columns.

The Fabric adapter inspects the warehouse's metadata to discover what it can serve. The CLI introspection commands work the same way against a Fabric data source as they do against Trino. Pass the configured source name:

pnpm lakeql-cli list-schemas --source reporting
pnpm lakeql-cli list-tables --source reporting --schema sales
pnpm lakeql-cli list-columns --source reporting --schema sales --table orders

Like the Trino adapter, the Fabric adapter builds its own SQL for metadata rather than calling a client convenience method: sys.databases for catalogs, and INFORMATION_SCHEMA.SCHEMATA / TABLES / COLUMNS for schemas, tables, views, and columns. The difference is one level down. The standalone trino-client package additionally exposes its own schemas()/tables()/views()/columns() convenience methods for callers using it directly outside the adapter. fabric-client has no equivalent.

Catalog listing filters out the fixed SQL Server system databases (master, tempdb, model, msdb). Schema listing filters out sys, INFORMATION_SCHEMA, guest, queryinsights, and _rsc, plus any schema matching the db_* role-schema pattern (for example, db_owner), so only user-facing schemas are returned. Column discovery reports name and type, but the adapter always returns description: undefined. Trino maps a comment column; Fabric's column query has no equivalent mapped in.

Previous page

Connection

Next page

API Reference