follow up question https://docs.dagster.io/_apidocs/internals#dagster.core.instance.DagsterInstance says that

Users should not directly instantiate this class; it is instantiated by internal machinery when dagit and dagster-graphql load, based on the values in the dagster.yaml file in $DAGSTER_HOME

but

make_airflow_dag

accepts a

DagsterInstance

at https://github.com/dagster-io/dagster/blob/master/python_modules/libraries/dagster-airflow/dagster_airflow/factory.py#L240-L245 Is there prior art or recommendations here? Would you advise against supplying a

DagsterInstance

to

make_airflow_dag

?

that doc block should get updated

DagsterInstance.get()

is the recommended way which uses the default

$DAGSTER_HOME

+

dagster.yaml

scheme

ok thank you 👍