Creating tag rules in compute cluster environments

With tag rules, you can apply Apache Atlas classifications to your assets based on regex expressions or similarity to a set of values in a table.

To start applying tags, go to Profilers and select your data lake.
Go to Profilers > Data Compliance > Tag Rules.
Click + Create Tag Rule.
Name your tag rule and add a description to it in General Information.
Select the tags to be applied from the list of available tags synchronized from the list of Atlas classifications.
If you select a child tag, its parent tag is also automatically selected. By default, if the child tag is applied to a column, the table receives the parent tag.

Select your Data Pattern Type:

Option	Description
Regular Expression	You can upload a text file containing your regex expression or directly type it in the Configure Tag Rule page. The required format of the CSV file can be seen by clicking Download Sample Tag Rule. Continue in step 7.
Single Column File Upload	Upload a CSV file with values to be matched against the actual values in your tables. After uploading your file, continue with step 11.

Option

Description

Regular Expression

You can upload a text file containing your regex expression or directly type it in the Configure Tag Rule page. The required format of the CSV file can be seen by clicking Download Sample Tag Rule.

Continue in step 7.

Single Column File Upload

Upload a CSV file with values to be matched against the actual values in your tables. After uploading your file, continue with step 11.

Creating regular expression based tag rule:

Optional: Define your regular expression for the table name.
When using Column Level regex expressions, you can define multiple expression for both of the following:
- Column Name
- Column Values
note
Regular expressions matching the same type of entity (column name or value) have the OR logical relationship between them. When using multiple regular expressions of the same type (table name, column name or value), even if one of the regular expressions match, it is considered as a match.
Define the Column Value Weightage in percentage with the slider.
The remainder percentage is the column name weightage percentage. The results of the individual regex matches are weighted according to this setting before determining the final result confidence for applying the tag.

note
A correctly formatted file is automatically processed by Cloudera Data Catalog. All details will be filled in this case.

Tag rule testing:

Optional: You can make a sanity check of your tag rule in Test Tag Rule by uploading a sample dataset in CSV format.

note
A final test called "Dry Run" is still needed to be passed to enable your tag rule.
Review all your input before clicking Create Tag Rule.
1. Click Confirm to finalize your tag rule.
Your tag rule is created with Status Disabled() and the Test Status will be Test Pending.
Click > Dry Run.

The Dry Run Test pane opens.
Click Run to start an on-demand dry run profiling job on up to 10 tables from your data.

Your tag rule becomes VALIDATED after a successful dry run.
After the "Dry run" test was passed, click > Enable to start your using your tag rule on your live data.