Conditional rules combined with data filtering capabilities improve how data processing systems work by making them more precise and efficient. This feature lets you apply rules selectively to specific datasets and fields, giving you more relevant and useful insights. When you create a rule, you'll now have an additional step where you can define a row filter. This filter works like a SQL WHERE clause - it specifies which rows of data the rule should evaluate.
You can choose to evaluate the row filters based on basic and advanced conditions.
- Basic conditions can be used for simple evaluation scenarios. You can configure
a basic condition by selecting the value and the associated operator. The
initial value field acts as an alias for all the target fields. Any condition
configured in the Basic tab is converted into an expression in the
advanced tab.
- Select the combination of
<attribute><operator><enter value><value>and type a value that should match the condition. - Select the combination of
<attribute><operator><select field > <field>and select a field that should match the condition. In this case, you can compare the selected field with the field of the same data type. You can use this for all available data types.
- Select the combination of
- Advanced conditions can be used for more complex evaluation scenarios. Within the Advanced tab, you can configure an expression using numerical, string and logical operators. To configure row filters for rule evaluation:
- Click Add Evaluated Rows Filter.
- Choose to evaluate the rows using basic or advanced conditions.
- Click Add filter to add more conditions to the row filter and select between Match All (logical AND) or Match Any conditions (logical OR).
- Once the rows are filtered, apply pass conditions for the filtered data assets.
- Save the rule and initiate the run.
Note: All the fields added from the evaluated row filter have to be mapped to the reference dataset or field.
Evaluate row filters using SQL queries
You can evaluate row filters using either basic and advanced conditions, or by writing your own SQL queries. The SQL query option is only available if you've selected Target Assets through the SQL option.
- Click Add Evaluated Rows Filter.
- Enter your SQL query in the text box. You can write SQL queries using the available SQL functions, and correlate data across multiple tables directly.
- After the query is entered, the specified conditions are evaluated based on the input, and the corresponding rows are filtered accordingly.
- After successful rule evaluation, the evaluated count shows the total number of records filtered by the conditions defined in the row filter.
Auto generate expression for row filters using natural language
The expression for advanced conditions in the row filters can be generated using the AI assist option.
- In the Evaluate row filters section, click AI assist.
- In the description box, enter your logic in plain English. Ensure the details of the relevant datasets or fields are added to the logic.
- Click Generate. The application will generate an expression for the row filter based on the inputs.
- Click Apply to save the changes and filter only those rows that satisfy the condition.
For more information on generating these expressions, refer to Auto generate rule expression for pass conditions using AI.