LakeQL
Overview
  • Package Map
query-builder
  • Filtering
  • Sorting & Paging
  • API Reference
config
  • Runtime Config
  • API Reference
trino-client
  • Client Setup
  • Executing Queries
  • Streaming
  • Metadata
  • API Reference
fabric-client
  • Client Setup
  • Executing Queries
  • API Reference
GitHub
LakeQL
  1. Packages
  2. fabric-client
  3. Client Setup

On this page

  1. Authentication
  2. Retry
  3. Connection Lifecycle

Client Setup

Create and configure a FabricClient.

Construct a FabricClient with the endpoint, catalog, and authentication to create a client.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
import { FabricClient } from "@lakeql/fabric-client"

const client = new FabricClient({
  host: "your-endpoint.datawarehouse.fabric.microsoft.com",
  port: 1433,
  catalog: "my_warehouse",
  auth: {
    type: "oauth2",
    tenantId: "...",
    clientId: "...",
    clientSecret: "...",
  },
})

Authentication #

The client supports two auth types. Both resolve to an Azure AD access token:

1
2
3
4
5
6
// OAuth2 client credentials
auth: { type: "oauth2", tenantId: "...", clientId: "...", clientSecret: "..." }

// Static bearer token
auth: { type: "bearer", token: "..." }

The oauth2 form accepts an optional scope field, but it currently has no effect. The token is always acquired for the Fabric SQL scope (https://database.windows.net/.default).

Retry #

The constructor takes an optional second argument, a RetryConfig. It applies to each query:

1
2
const client = new FabricClient(props, { maxRetries: 3 })

Connection Lifecycle #

The client manages a TDS connection pool internally. When you are done, call close to release it:

1
2
await client.close()

A closed client throws FabricClientError if used again.

Previous page

fabric-client

Next page

Executing Queries