dbt2looker

Use dbt2looker to generate Looker view files automatically from dbt models.

Want a deeper integration between dbt and your BI tool? You should also checkout Lightdash - the open source alternative to Looker

Features

• Column descriptions synced to looker
• Dimension for each column in dbt model
• Dimension groups for datetime/timestamp/date columns
• Measures defined through dbt column metadata see below
• Looker types
• Warehouses: BigQuery, Snowflake, Redshift (postgres to come)

Quickstart

Run dbt2looker in the root of your dbt project after compiling looker docs.

Generate Looker view files for all models:

dbt docs generate
dbt2looker

Generate Looker view files for all models tagged prod

dbt2looker --tag prod

Install

Install from PyPi repository

Install from pypi into a fresh virtual environment.

# Create virtual env
python3.7 -m venv dbt2looker-venv
source dbt2looker-venv/bin/activate

# Install
pip install dbt2looker

# Run
dbt2looker

Build from source

Requires poetry and python >=3.7

For development, it is recommended to use python 3.7:

# Ensure you're using 3.7
poetry env use 3.7  
# alternative: poetry env use /usr/local/opt/[email protected]/bin/python3

# Install dependencies and main package
poetry install

# Run dbtlooker in poetry environment
poetry run dbt2looker

Defining measures

You can define looker measures in your dbt schema.yml files. For example:

models:
  - name: pages
    columns:
      - name: url
        description: "Page url"
      - name: event_id
        description: unique event id for page view
        meta:
           measures:
             page_views:
               type: count