Create Parquet file as stream
Creates a Parquet file from a DataReader or DataTable and returns the result as a Stream.
Example
Using data table as Source:
Example 
This Flow performs an incremental load of a dimension into a Parquet file stored in S3. It first retrieves the list of S3 objects and uses a custom function to check whether UC038_Dimension.parquet already exists. The dimension data is then read as a DataReader and processed in chunks, with each chunk converted to Parquet format as a stream. For each chunk, the Flow evaluates the file existence flag: if the Parquet file is already present, the stream is appended to the existing object; otherwise, a new Parquet object is created and uploaded. This pattern is useful for implementing incremental data loads without recreating the entire Parquet file.
Properties
| Name | Type | Description |
|---|---|---|
| Title | Optional | The title for the create action. |
| Source | Required | Specifies the source of the data, which can either be a DataReader or DataTable. |
| Result variable name | Required | The name of the variable you can use to reference the Stream in other actions. |
| Description | Optional | Additional notes or comments about the action or configuration. |
Returns
