Jet Analytics Parquet Data Source

The Jet Analytics Parquet data source connects to Parquet files stored in a variety of locations, including local drives, Azure Blob Storage, AWS S3, SharePoint/OneDrive, Google Cloud Storage, and SFTP. It can read files across folders and subfolders and supports aggregating multiple files into a single table.

For general connection settings shared across all Jet Analytics Enhanced data sources, see Data Sources General Connection Settings.

Comma-separated parameter syntax

For any setting that accepts a comma-separated list, individual items can be wrapped in square brackets to allow commas within that item. For example, the following is parsed as a three-item list:

yyyyMMdd,[dd,MM,yyyy],yyyy-MM-dd

Connection settings

Path: The path to the Parquet file or folder. Only one path can be specified per data source connection. For SharePoint/OneDrive, a forward slash (/) represents the root of the Drive. For other non-local locations, the path format may differ depending on the location type.
Include sub-folders: When enabled, Parquet files in subdirectories are also collected.
Note:
This setting is ignored for AWS, Azure, and Google Cloud locations.
Included file types: A comma-separated list of file extensions to include. Files with extensions not in this list are ignored.
File aggregation pattern: A comma-separated list of file name patterns (supporting * and ? wildcards). All files matching a pattern are treated as a single table. Files grouped by pattern are expected to share the same schema — if they do not, the full data load will fail.
Location: The storage location type. Determines which additional fields are shown. See the location-specific sections below.

Location-specific settings

Local file or folder

Use this option when the file is stored on a local drive accessible from the machine running the Jet Analytics Ingest Service.

To connect to an Azure File Share, use the UNC path format:

\\<storageAccountName>.file.core.windows.net\<fileShareName>\<folder>\<filename>.parquet

Before using this path, mount the file share as a network drive. In the Azure portal, navigate to the file share, click Connect, click Show Script, and copy the PowerShell script. Run the script in PowerShell as the same user account that runs the Jet Analytics Ingest Service — the script will not work if run as Local System.

To run PowerShell as the Ingest Service user: search for PowerShell, right-click it, select Open file location, then right-click the application and select Run as a different user. Paste and run the script. A confirmation message appears when the drive is mounted.

Azure Blob Storage

Azure blob connection string: The complete connection string to the container. See Configure a connection string (Microsoft Learn).
Azure Tenant ID: Required for OAuth 2.0 client grant authentication.
Azure Client ID: The application client ID for OAuth 2.0 authentication.
Azure Client Secret: The client secret for the application.
Azure storage account name: The name of the storage account hosting the container.
Azure blob container: The container name.

AWS S3 Bucket

Note:

Only Access Key ID authentication is currently supported for AWS S3.

AWS region: The region string for the bucket (for example, eu-west-2).
AWS access key ID: The access key ID for the S3 bucket.
AWS secret access key: The secret access key for the S3 bucket.
AWS bucket: The bucket name.

SharePoint or OneDrive

SharePoint/OneDrive Client ID: The Client ID of the app registration. Requires Files.Read.All and Sites.Read.All Application permissions. Adding Group.Read.All and User.Read.All is recommended for broader access.
SharePoint/OneDrive Client Secret: The client secret for the app registration.
SharePoint/OneDrive Tenant ID: The Tenant ID of the organisation hosting the SharePoint site.
SharePoint/OneDrive Drive ID: The Drive ID of the Document Library to connect to. See Locate a SharePoint/OneDrive Drive ID.

Google Cloud Storage