Cloud Dataprep does not modify a source. Instead, a set of metadata is associated with the raw data, which enables the storage and transformation of the source. On export, a new version of the data is written to the preferred output destination. For information on the formats supported for input, see Supported File Formats.
Imported and Wrangled Datasets
- When data is imported, it is stored by the platform as an imported dataset. This reference object is not modified.
- A wrangled dataset can be created from an imported dataset. A wrangled dataset associates an imported dataset with a recipe to modify the data. See Dataset Types for more information.
- You modify the recipe associated with a dataset to transform imported data.
- From ___, click Datasets. then click Import Data. [TODO:(amancuso): add URLa]s/screenshots; these steps may make a quickstart]
- To begin working with a dataset, you must first import it into a flow, which is a container for datasets. Click the Add Dataset to a Flow checkbox, and enter the name for a new flow.
- Click Import & Add to Flow
- After the flow is created, open the flow and select the dataset to begin transforming it (see Transform Basics). [TODO(amancuso): the last two steps need clarification, validatiom]