Skip to content

Configure a development environment

This guide configures the checked-in bricksgdpr dbt profile to use Databricks OAuth and installs the repository's pinned Python dependencies.

Prerequisites

You need:

  • Databricks CLI 1.7.0 or newer;
  • uv;
  • access to a Databricks workspace; and
  • the repository checked out locally.

Install jq as well if you will provision or validate persona identities.

Create the local environment file

From the repository root:

cp .env.example .env

Fill every blank workspace/account value needed by the workflows you intend to run. The example contains no real workspace host, warehouse ID, account ID, group ID, or operator email.

Warning

Do not overwrite an existing .env until you have reviewed its target. The file is ignored by Git and can contain machine-specific identifiers, but it must never contain the pseudonymization pepper or a Databricks token.

Review these dbt connection values in .env:

DATABRICKS_HOST
DATABRICKS_HTTP_PATH
DBT_CONTROL_CATALOG
DBT_CONTROL_SCHEMA
DBT_PROJECT_CATALOG
DBT_SCHEMA_PREFIX

The checked-in profiles.yml reads them and uses auth_type: oauth.

Authenticate the CLI profile

databricks auth login \
  --host "$DATABRICKS_EXPECTED_HOST" \
  --profile bricksgdpr

Confirm that the profile resolves successfully:

databricks auth describe --profile bricksgdpr --output json

The response should report a successful authentication for the expected workspace host.

Load the variables

set -a
source .env
set +a

This exports the values for both dbt and the repository scripts in the current shell.

Install and parse

uv sync --all-extras
uv run dbt parse

uv sync --all-extras installs the pinned dbt-core and dbt-databricks versions plus ruff and ty. A successful parse confirms that dbt can load the project configuration and write its local artifacts.

Note

dbt parse does not create the Unity Catalog catalog, secret, groups, or model relations. Complete Deploy the demo for the first time for a new workspace.

Consult Project configuration for every environment variable and default.