The cloud-native orchestrator for the whole development lifecycle, with integrated lineage and observability.

dagster

Hello, I'm looking for moving my Airflow DAGs to Dagster Assets/Jobs, and I'm wondering if there is something like Airflow `PostgresHook(conn_id='db').get_pandas_df(query)` in Dagster?

we currently don’t have postgrest-pandas integration.

this can be done via resources or I/O manager. for example, we have native snowflake-pandas integration (<https://docs.dagster.io/integrations/snowflake/reference#using-pandas-and-pyspark-dataframes-with-snowflake|doc>) where you can write an asset whose input or output is pandas dataframe, the <https://docs.dagster.io/concepts/io-management/io-managers|I/O manager> can read and write df from/to snowflake.

here’s the concept mapping from airflow to dagster: <https://docs.dagster.io/integrations/airflow#airflow-vs-dagster-concept-map>

for your case, you may do something like, where you customize the `get_pandas_df`:
```from dagster import asset, Definitions, ConfigurableResource

class MyDBResource(ConfigurableResource):
    conn_id: str
    def get_pandas_df(self) -&gt; Response:
        ...

@asset
def data_from_db(my_conn: MyDBResource) -&gt; Dict[str, Any]:
    return my_conn.get_pandas_df(query)

defs = Definitions(
    assets=[data_from_db],
    resources={
        "my_conn": MyDBResource(conn_id="..."),
    },
)```