We are very happy to announce the release of the snowplow-web v.0.2.0 dbt package.
For the latest installation instructions please refer to dbt’s hub.
This release predominantly brings support for the BigQuery adapter.
Improvements to the package architecture have also been made by reducing the number of models. In order to achieve this, breaking changes were required.
Breaking Changes
- Retire the
snowplow_web_users_manifest
model (#7)
Features
Improvements
- Update Redshift integration test data (#8)
- Materialize scratch models as views (#13)
- Remove dbt_date dependency (#18)
Fixes
- Update datediff based filters to timestamp based (#11)
- Fix filters to determine sessions_this_run (#12)
- Remove dbt_utils package import (#17)
- Fix package install instruction in custom_example directory (#10)
Upgrading
In order to retire the snowplow_web_users_manifest
model, additional columns needed to be added to the snowplow_web_base_sessions_lifecycle_manifest
incremental model. As such, in order to upgrade all tables in the snowplow_manifest
schema will need to be dropped alongside a full refresh of all models. This can be achieved with:
dbt run --models snowplow_web tag:snowplow_web_incremental --full-refresh --vars 'teardown_all: true'
This will drop all models within the package as well as any custom modules you have written and then replay all events.
More information
- Checkout the snowplow-web package docs for a quickstart guide as well as an explanation of operating and configuring the package.
- Checkout the web data model section of Snowplow Docs for more information on the web models structure.