Function: applyLakebaseEnv()

applyLakebaseEnv(config?, signal?): Promise<{ resolved: LakebaseConnection; user?: string; }>

Resolve the connection AND apply every Postgres env var a Lakebase pool needs, returning the resolved connection plus the username that was applied.

This is the whole set, which is the point of having one function for it: createLakebasePool() throws unless LAKEBASE_ENDPOINT, PGHOST, PGDATABASE, and a username (PGUSER, or DATABRICKS_CLIENT_ID for a service principal) are all present, and a Databricks App postgres resource binding supplies only the first. Anything that wants a working pool - the lakebase plugin, or AppKit’s PERSISTENT cache, which quietly degrades to in-memory when the pool cannot be built - needs all four, so callers should not pair applyLakebaseToEnv with their own username lookup.

PGUSER is applied with ??= like the rest, so an explicitly configured value stays authoritative. The lookup returns undefined rather than throwing when it cannot determine a user, leaving the pool to resolve its own.

Parameters

config?

LakebaseResolverInputs

signal?

AbortSignal

Returns

Promise<{ resolved: LakebaseConnection; user?: string; }>

Example

import { lakebaseResolver } from "@dbx-tools/appkit";

// Enough for `createLakebasePool()` to build a pool.
const { user } = await lakebaseResolver.applyLakebaseEnv({ autoCreate: false });