@Jeremiah would be so useful! Currently, it's a blocker for us to use Prefect instead of Airflow. To make it clear features that block us:
• to be able to see a full graph of all flows
• to be able to run a flow with all its ancestors. A typical use case - we update one of the data sources and we need to run all flows that use that data source or derived data sources
The following is inconvenient but we can work around it:
• register all flows in a folder including subfolders
• a simpler way to register an upstream flow. Currently code becomes too unreadable when I have to define a flow itself, then define another flow as StartRunFlow, then define StartRunFlow for each of the dependencies, and finally set another flow that actually contains the relationships. Instead of all the above would be great to have it as:
â—¦ define a flow itself
â—¦ set flow.set_upstream(a list of dependencies aka flows)