Learn how to use Looker (Google Cloud core) to automatically generate a basic data model from a connection that is included in each instance.
Each Looker (Google Cloud core) instance comes with the sample LookML project installed by default on the instance. The sample LookML project includes curated LookML models that are based on the public BigQuery dataset the Look eCommerce
. The sample LookML project uses a connection, sample_bigquery_connection
. You can use this connection to generate a new data model that is based on one of the sample project's tables. Since this quickstart is for instructional purposes only, you'll delete the data model at the end of the process.
Before you begin
To follow along with this quickstart, you'll need the following:
- Access to a Looker (Google Cloud core) instance that includes the sample LookML project.
- Someone with the Looker Admin IAM role must enable the BigQuery API for the Google Cloud project that contains your Looker (Google Cloud core) instance.
- You must have either the Looker Admin role or the
develop
permission.
Navigate to the Create a Model page
- On the Looker (Google Cloud core) homepage, enable Development Mode by selecting the Development Mode toggle in the main navigation menu.
- On the Looker (Google Cloud core) homepage, click the Create button in the main navigation menu to open the drop-down menu.
- From the drop-down menu, select LookML Model to open the Create a Model page.
Define your model
On the Create a Model page, follow these steps:
- In the Select Database Connection section, select
sample_bigquery_connection
from the drop-down menu. - Select the radio button to add the model to the
sample_thelook_ecommerce
project, and click Next. - In the Select Tables section, leave the default Google Cloud project in the Select GCP project field.
- In the Datasets field, select
bq_dataset
. - Click the arrow to the right of
bq_dataset
to get to the Tables field. - In the Tables field, select
bq_table
. - Click the arrow to the right of
bq_table
to get to the Fields field. - In the Fields field, select all the fields, and click Next.
- In the Select Primary Keys section, select Name as the primary key, and click Next.
- In the Select Explores to Create section, select the
bq_table.view
file, and click Next. - In the Enter Model Name section, enter a unique model name.
- Click the Complete and View Model button. This action takes you to the Looker IDE.
View your model
Once you've completed all fields in the Create a Model page, view the model that Looker generated.
View the model files in the Looker IDE
- After clicking the Complete and View Model button, you will be in the Looker IDE. In the IDE file browser, you see the folders for the pre-existing sample LookML models, which are
0_start_here
,1_basic_lookml
,2_intermediate_lookml
, and3_advanced_lookml
. You also see the folders created for the generated model,models
andviews
. - Expand the
models
andviews
folders to see the LookML files that Looker created for the model that you defined on the Create a Model page. The blue dot next to each file indicates that they are new files and not yet deployed to production.- The
.model
file defines the model. In the model file, you can see the connection that you selected, thebq_table
view, as well as other default configurations. - The
bq_table.view
file defines the view. In the view file, you can see thebirthdate
,name
, andrating
fields that you selected and how they are defined in the BigQuery database table, as well as a defaultcount
measure that Looker added.
- The
- You also see the
README.md
file that Looker generated because the Generate a Readme file that gives more information about using LookML checkbox on the Create a Model page was selected by default.
View the model on the LookML Projects page
- From the IDE, open the main navigation menu by clicking the Main menu icon.
- If you aren't already in the Develop menu, click Develop.
- From the Develop menu, select Projects to navigate to the LookML Projects page.
- View your model in the Models column of the sample_thelook_ecommerce row. It appears along with the
advanced_ecomm
,basic_ecomm
, andintermediate_ecomm
models from the sample LookML project.
View the Explore that was created by the model
- From the Projects page, click the name of your model to return to the model file in the IDE.
- With the model file open in the IDE, click the See file actions arrow next to the filename at the top of the IDE.
- Select Explore Bq Table- your_model_name from the drop-down menu. This selection takes you to the Bq Table Explore that was created by your model.
- In the field picker, note the Birthdate Date, Name, and Rating dimensions that you selected on this Create a Model page, as well as the default Count measure.
The model files, model configuration, and Explore that you've viewed are available only in your personal Git branch while in Development Mode. If you wanted to make this model available to other users in your Looker (Google Cloud core) instance, you would commit your changes, merge your branch, and deploy the production branch. However, for the purposes of this quickstart, you will delete this model, as described in the next section.
Clean up
To maximize Looker performance, unnecessary models shouldn't be deployed to production. To maintain a clean personal Git branch and avoid accidentally deploying this model to production, complete the following steps to delete the model:
- From the Explore, click the Go to LookML link in the field picker. This action takes you to the Looker IDE.
- In IDE, select the Git Actions icon.
- In the Git Actions menu, click Revert to.
In the Revert to window, make sure that the Revert uncommitted changes option is selected, and click Confirm.
In the IDE, you should no longer see your model files.
To delete the model configuration, navigate to the Projects page by clicking Projects in the Develop menu.
Locate your model in the table, and click the Configure button in its row.
Make sure you are in the window for your model by checking the name in the Model field, and click Delete Model Configuration.
What's next
- Generating a LookML model
- Looker IDE overview
- Managing LookML files and folders
- Common LookML patterns
- Using version control and deploying