​

QuickFlows

A QuickFlow binds source and destination connections to a cluster and records how VegaFlow selects and moves data. Validate each connection and bring the cluster to a usable state before relying on the flow.

​

Selection modes

Stream and column selection use three modes:

• all: include/exclude lists must be empty.
• include: the include list must contain non-empty unique identifiers; the exclude list must be empty.
• exclude: the exclude list must contain non-empty unique identifiers; the include list must be empty.

Use catalog discovery from the selected connection to choose real identifiers. Column rules can be applied per selected stream where the schema supports them.

​

Source modes

​

Snapshot

Snapshot reads the selected source as a complete point-in-time data set. Cursor strategy, cursor fields, resume token, and initial-snapshot controls are omitted.

​

Incremental cursor

Incremental cursor mode requires at least one cursor. A single cursor can use the default single strategy; multiple cursors require an explicit strategy. Cursor fields cannot be blank, ordering must be valid, and tie-breaker settings should produce deterministic progress. Some connector/runtime paths may support resume-token semantics. Use only the mode and fields present in the current connector schema and generated API contract.

​

Schema evolution

When schema evolution is configured, its enabled value is required. Enable it only when changes from the source should be accepted by the destination workflow; otherwise keep destination expectations explicit.

​

Lifecycle and permissions

Create, inspect, update, and delete QuickFlows through collection or instance permissions. Before an update, consider active schedules and how new connection, cluster, selection, or cursor settings affect the next run. Before deletion, remove or retire schedules that depend on the flow. Continue with runs and schedules.