For Foundry to connect to the Palantir Foundry Connector 2.0 for SAP Applications ("Connector"), a source needs to be configured in the Data Connection app. The configuration requires an agent, user credentials for the connection, and the URL for the SAP Source System.
To create an SAP source follow these steps:
Navigate the Data Connection app in the sidebar.
Select Add Custom Source option.
Choose the connection Via an agent option.
Give the source a name and location.
Complete the Source Setup requirements:
Copied!1 2 3
type: magritte-sap-source url: https://<host>:<port>/sap/palantir usernamePassword: <username>:{{password}}
The expression {{password}}
should be written as it appears. You can then enter the password under Encrypted values on the right of the dialog box and it will be encrypted.
Save the source definition.
These are the extra parameters that can be configured on the source.
Parameter | Required? | Default | Description |
---|---|---|---|
url | Y | The base URL of the SAP add-on service endpoint. | |
usernamePassword | Y | The username and password are those for the technical user that was created for Foundry to use to connect to SAP. | |
useKernelJsonSerialization | N | false | Switch on kernel JSON serialization for paginated data returned from SAP. Only works for JSON formatted data: checks will fail if set to true when useTsvFormat:true . |
useTsvFormat | N | false | Switch on using TSV format for paginated data returned from SAP (vs JSON). |
output | N | 50,000 rows | Definition of the maximum file size (rows or bytes) that will be returned from SAP. |
convertDatesToStrings | N | false | Ingest all dates as strings for this source. |
proxy | N | None | Proxy configurations for connecting to SAP. |
cacheConfigurations | N | None | Cache configurations for configuring cache timeouts for different SAP Object types. |
The maximum file size of each Parquet file in Foundry datasets can be defined on a specific sync (for that specific sync) or on the Source (for all syncs).
If you would like to change the maximum size of each Parquet file for all syncs under a given source, you can configure this in the source config.
Examples:
Copied!1 2 3 4 5
output: maxFileSize: type: rows rows: max: 10000
Copied!1 2 3 4 5
output: maxFileSize: type: bytes bytes: approximateMax: 400MB
Specifying the maximum size in bytes is only approximate. The resulting file sizes might be slightly smaller or larger.
If specifying a maximum size in bytes, the number of bytes needs to be at least twice as large as the in-memory buffer size of the Parquet writer (which defaults to 128MB).