Include column + metric UUIDs in dataset exports (Preset UI / Preset CLI / Superset export endpoints)

What is the problem or goal you're trying to solve or accomplish?

Customers want to programmatically manage dataset metadata (specifically Folders for columns/metrics) in Apache Superset / Preset.

To create folders and assign columns/metrics to them via the Update Dataset (PUT) API, they need the UUID for each column and metric. However, when exporting datasets (via UI export, Preset CLI export, or the dataset export endpoint), the export artifact does not include the UUIDs for columns/metrics, so the export can’t be used as a self-contained source-of-truth for automation/migrations.

How are you solving it currently?

They must do a multi-step workaround per dataset:

• Call GET “Get a Dataset” for each dataset ID to retrieve column UUIDs + metric UUIDs.

• Build a script to collect those UUIDs.

• Call PUT “Update a Dataset” to create a folder UUID and then associate the collected column/metric UUIDs to that folder.
  This adds extra API calls, scripting complexity, and makes exports less useful for GitOps-style workflows because the exported file alone is missing the needed identifiers.

What is your recommended solution?

Update dataset export outputs so they include the UUIDs for each column and metric.

Scope should cover all export paths (UI export, Preset CLI export, and the dataset export endpoint).
Optional (nice-to-have): include folder metadata/associations as well, but the minimum requirement is that column/metric UUIDs are present in the export, so customers don’t need an additional GET call per dataset just to recover identifiers.