Table of Contents

Search

  1. Preface
  2. Introduction to Data Validation Option
  3. Repositories
  4. XML Data Source
  5. Tests for XML Data Sources
  6. Connections
  7. Expressions
  8. Table Pairs
  9. Tests for Table Pairs
  10. Single-Table Constraints
  11. Tests for Single-Table Constraints
  12. Examples of Tests from Spreadsheets
  13. SQL Views
  14. Lookup Views
  15. Join Views
  16. Aggregate Views
  17. Business Intelligence and Reporting Tools Reports
  18. Dashboards
  19. DVOCmd Command Line Program
  20. Troubleshooting
  21. Appendix A: Datatype Reference
  22. Appendix B: Reporting Views
  23. Appendix C: Metadata Import Syntax
  24. Appendix D: Jasper Reports
  25. Appendix E: Glossary

Data Validation Option User Guide

Data Validation Option User Guide

Bad Records for a Table-Pair Test

Bad Records for a Table-Pair Test

Bad records are the records that fail a value or a set test.
When you select a test on the Tests tab, the bad records appear in the
Bad Records
view.
The Bad Records view displays a delimiter before and after each string value to help you identify whitespace in a record of string data type. The Bad Records view displays the difference in value along with the percentage of difference in a record of a numeric data type. Data Validation Option uses the following formula to calculate the percentage of difference:
(<larger value> - <smaller value>) ÷ <larger value> * 100
The columns that appear on the
Bad Records
view differ based on the test.
Aggregate Tests
Aggregate tests do not display bad records. The
Results
view displays the test result value from Field A, the test result value from Field B, and the comparison operator.
Value Tests
Value tests display the following columns for each bad record:
  • The key for Table A
  • The field or expression from Table A being compared
  • The key for Table B
  • The field or expression from Table B being compared
Set Tests
Set tests display the following columns for each bad record:
  • The result from Table A
  • The result from Table B
If you choose to view more details about numerical bad data records, the tests display the following additional columns:
  • Difference
  • % Difference
If you compare two fields where one field has a value and the other is empty, Data Validation option considers the record as a bad record. If the repository is on Oracle, the database stores the empty field as NULL. A bad record with NULL value in an Oracle repository can be either NULL or an empty field.

0 COMMENTS

We’d like to hear from you!