Import

All your Hashboard models, metrics and data connections are available within Hashquery. You can see an overview of your project by simply printing it:

from hashquery import project
print(project)

Currently, you can only pull content from Hashboard. In future versions, you’ll be able to push content created in Hashquery back into Hashboard.

Models

Models can be accessed using project.models.my_model_alias.

Models in Hashboard are meant to be 1:1 with Models in Hashquery. At this phase in development however, there are some known differences:

Models in Hashboard with Custom SQL attributes or measures may use interpolation ({{ }}) and disambiguation syntax (alias.) that isn’t valid inside of all queries that Hashquery generates.
Models in Hashboard store additional metadata, such as column descriptions, cache TTLs, change history, dbt metadata, and others. Hashquery does not yet support all these options.
Hashboard and Hashquery use separate caches.

Metrics

Metrics can be accessed using project.metrics.my_metric_alias.

Metrics are imported into Hashboard as a Model. When run, they will fetch the underlying data for their sparkline.

At this time, some aspects of metrics are not fully supported when imported.

Metric filters are not consistently supported, and may be silently ignored.
Hashboard and Hashquery use separate caches.

Connections

Connections can be accessed using project.connections.my_connection_name. Connections can be passed into the first parameter of Model.with_source().

The name of the connection is a lowercased version of the name as it appears in the Hashboard UI, with spaces converted to underscores. You can print(project.connections) to see all the names if you need help in finding the name for your connection.

Connections are just a reference; all processing occurs on Hashboard servers and no database credentials are sent to Hashquery clients.