Execution & sharing — quiz

Which is reusable across workflows and can expose its own settings?

1. Metanode
2. Component
3. Annotation
4. Workflow group

Answer: Component — Metanodes are purely organisational.

A good node label says:

1. The node type
2. What the node is for in this workflow
3. The row count
4. The author

Answer: What the node is for in this workflow — The type is already visible from the icon.

Which node gives a per-column profile with missing counts and distributions?

1. Statistics only
2. Data Explorer
3. Table View
4. GroupBy

Answer: Data Explorer — The equivalent of df.info() plus df.describe().

The most reliable way to catch a workflow producing wrong numbers:

1. Check the output only
2. Check the row count after every node that can change it
3. Add more annotations
4. Use components

Answer: Check the row count after every node that can change it — Otherwise you are bisecting eighty nodes at the end.

Why reset a workflow before exporting?

1. It is required
2. It strips cached data, shrinking the file and avoiding sending data
3. It fixes errors
4. It updates nodes

Answer: It strips cached data, shrinking the file and avoiding sending data — Exported workflows otherwise carry their cached tables.

The free way to run a workflow on a schedule:

1. Business Hub
2. The batch executor driven by cron or Task Scheduler
3. Manual execution
4. Components

Answer: The batch executor driven by cron or Task Scheduler — Business Hub is the commercial option.