Trigger_Python
Node
Node Description
Execute downstream flow
Retrieves table(s) from Ganymede data lake, processes via user-defined Python function, and uses result to execute downstream flow
User-Defined Python
Processes data, then triggers downstream flow
Parameters
- df_sql_result :
pd.DataFrame | list[pd.DataFrame]
- Table(s) or list of tables retrieved from user-defined SQL query
- ganymede_context :
Optional[GanymedeContext]
- Ganymede context variable, which stores flow run metadata
Returns
tuple[str, FlowInputs]
Tuple where first element is the ID of the flow to trigger
and second element is the FlowInputs class containing all inputs for the flow to trigger
Notes
The ID for the triggered flow can be found by navigating to the flow in the Ganymede UI and clicking on Manage. The ID is an immutable string found on the Configuration panel.