Use a more maintained
databricks-labs-pytesterplugin for an extensive fixture set.
Pytest plugin for testing dbx-based projects.
To use the fixtures in your project, simply add pytest-dbx to your project's dev-dependencies
(where you would place all your test and build dependencies).
In unit tests you can use the dbx_spark fixture to have a spark session available as in Databricks
def test_function(dbx_spark):
sdf = dbx_spark.createDataFrame([[1], [2], [3]], ['a'])
assert sdf.count() == 3The dbutils_fixture and mlflow_local are automatically used.