Specifies the number of rows the provider scans when dynamically determining table columns.

Data Type

string

Default Value

"100"

Remarks

When this property is set, the server scans the specified number of rows (objects) in the JSON document to detect columns and data types. The scan follows nested objects, counting each object array as a single row.

If no schema (RSD) file is available for a table, columns are determined dynamically using this scan. This behavior applies when using GenerateSchemaFiles or when a schema file is not present.

Higher values improve detection accuracy but may increase request time. Setting this property to 0 causes the server to parse the entire JSON document.

This property is useful for controlling the trade-off between schema detection accuracy and query performance when parsing JSON data.

See Also

• Automatic Schema Discovery: Shows how to configure column discovery using RowScanDepth, FlattenObjects, and FlattenArrays.
• Modeling JSON Data: Introduces the different options for modeling data in the server.