RPI performs validation to ensure that records within the file to be loaded conform to the rules specified within the data project’s definition.
The first [n] records from the file are validated, followed by every 1 record in [x]. [n] is defined by system configuration setting NumberFirstLinesToRead. [x] is defined by system configuration setting ReadEveryXLines.
A number of types of validation are performed:
•Row-level validation: RPI must be able to parse the row. Validation fails if too many or too few fields are present.
•Data type validation: the validation checks undertaken are determined by the type of data:
o String: the data must fit the defined size
o Integer: the data must be a valid integer, and must fit the defined size
o Decimal: the data must be a valid decimal or integer, and must fit the defined size
o Date: the data must be a valid Date, and no Time component may be present
o Time: the data must be a valid Time, and no Date component may be present
o DateTime: the data must be a valid DateTime, Date or Time.
•Key constraint validation: a record must not violate any user-specified primary key constraint.