Configure matching rules for validation objects

This topic describes the background information, limitations, and configuration methods of matching rules for validation objects.

Background information

When you create a data validation task, you can specify the validation objects by specifying objects, importing a CSV file, or configuring matching rules. When you configure matching rules, you can specify the validation objects by using wildcard rules and configure the object mapping logic between the source and the target. This way, you can easily and efficiently match the validation objects, thereby reducing the configuration complexity when you have a large number of validation objects.

The following table describes the wildcard rules supported for data validation.

The following rules apply to wildcard rules:

• The target database name and table name cannot both use wildcard rules. For example, kd_test*.person*=kd_test*.person*.

• If both the source and target databases use wildcard rules, the database-level expressions must be exactly the same.

• If both the source and target tables use wildcard rules, the table-level expressions must be exactly the same.

• If the target database uses a wildcard rule, the source database must use a wildcard rule.

• If the target table uses a wildcard rule, the source table must use a wildcard rule.

Limitations

• You can specify multiple rules for data validation. However, each rule must not contain line breaks or spaces.

• The validation object rule cannot be empty, but the object exclusion rules can be empty.

• When you select a validation object by using a matching rule, the table name cannot contain special characters (spaces, line breaks, or .|"'`()=;/&*?[][!]).

• Data validation does not support mapping tables in the same database at the source to different databases at the target by using multiple matching rules. For example, a.a* = b.a* & a.b* = c.b*.

Considerations

• After you configure the validation object rules and the object exclusion rules, if the source table name is in the difference set between the validation object rule and the object exclusion rules, the object can be selected.

  Note

  The difference set is a new set that contains all elements that exist in the first set but do not exist in the second set.

• If a table is configured with a renaming mapping rule, the renaming is prioritized. For example, if both a.b[0-3] and a.b[3-5]=a.c exist, the table a.b3 is renamed to a.c.

Configure a matching rule

1. Create a data validation task and configure it to the Select Type & Objects step.

   For more information, see Create a data validation task.

2. In the Select Type & Objects step, select the validation type and validation object.

   You can select validation objects through two methods: Specify Objects and Match by Rule. This topic describes how to configure a matching rule.

   1. In the Validation Objects section, select Match by Rule.

   2. Enter Validation Object Rule and Object Exclusion Rules (optional). For more information about supported wildcard patterns, see Wildcards.

   3. Click Verify.

      After the validation is completed, you can click Preview to view the matching result. The wildcard validation object rules and the object exclusion rules that you specify apply only to tables. Matching Results contains the final objects, added objects, and reduced objects.

      Objects	Description
      Final	The validation object that is finally matched by the configured matching rule.
      Added	The validation object that is newly added in the final matching result compared with the matching result configured in the previous step.
      Reduced	The validation object that is reduced in the final matching result compared with the matching result configured in the previous step.

3. Complete the subsequent task configuration as prompted.