We do not perform data validation with rules from DQM. We use self-implemented scripts (e.g. SQL) for data checking. Is there a way to still include the results of this data check in DQM so that users can use the existing functionality to quickly navigate to and clean up erroneous objects?
Yes, this is possible.
Step 1: Provide the defects for DQM
For the integration of defects from external data checks, the defects must be stored in a table. The name of the table can be freely selected. It is possible to use different tables for different data checks.
The table must contain reference information about a defective UT object. These are:
- UT Object ID: OBJ_ID
- UT Object class ID: ACLASS_ID oder CLASS_ID
- UT Object class subtyp: SUBTYPE_ID oder UTCLSUB_ID
Optional is a field for a message text. This can provide information about the type of defect. If the data check reveals different defects, this is extremely useful, so that one ultimately also knows what is faulty.
Step 2: Integration of the results into DQM
The provided table can be evaluated using the "Load defect UT objects from table" rule. The name of the table must be specified in the rule setting in the "UT Object Key Table" rule setting. If the current user is not the owner of the table, the owner must be specified in the "Table Owner" rule setting.
In the "Error Message Column" rule setting, the field for the message text is specified, if one exists. If nothing is specified here, the specified field does not exist or no message text is found in the field for a defect, the message text stored in the "Error Message" rule setting is used.
The results can be saved and reopened at a later time. Furthermore, saving as a work order (error cleanup list) is possible. This allows errors to be successively cleaned up by users without having to perform a new data check.
Unlike other UT object rules, the "Load defect UT objects from table" rule does not evaluate the set checkload pipes, but returns the complete contents of the table.