3516476 - Field length not correctly detected when creating Data Flow from file-based sources in SAP Datasphere

When creating a Data Flow in SAP Datasphere based on an Excel file, the system may show the wrong field length.

Example: A field can be assigned String(2) even though some values in the dataset need String(3) or longer.

SAP Datasphere

1. Logon to Datasphere tenant
2. Go to Data Builder
3. Create a Data Flow from file-based data source
4. Select an Excel file
5. Check the Data Type of some Field (e.g. named columnA) in Columns section
   => The Data Type should be String(3) but it shows as String(2) which is incorrect

• This behavior is by design.
• Excel, CSV, and JSON files do not provide metadata about field definitions.
• Therefore, SAP Datasphere infers the schema by checking only a subset of the data (see Point 5 in SAP Datasphere Document Add a Source to a Data Flow).
• If longer values appear outside of the scanned subset, they are not considered, and the suggested Data Type may be too short.

Workaround 1: Manually adjust the column definition so it covers the full length of the values in the dataset.

“For CSV, JSON, or Excel files, the schema is calculated based on a subset of data and hence you can adjust/verify the suggested data as per your requirements to cover all the data.”

Workaround 2: Manually change the Data Type of the field to the correct definition.

Add a Source to a Data Flow | SAP Help Portal

SAP Datasphere, Data Flow, Field Length, Data Type String, SFTP, Upload, Metadata, CSV, JSON, Excel, Schema, Adjust Column Definition. , KBA , DS-DI-DF , Data Flows , CA-DI-EMB , Data Intelligence Embedded in Data Warehouse Cloud , Problem