The SQL Pattern Matching rule ensures values in a dataset match specified text patterns using SQL LIKE syntax.
This rule is commonly used to:
- Identify and validate values like email domains, phone number formats, or prefixes.
- Enforce standardized value formats within columns for consistency.
Example Usage:
Check if email addresses belong to Gmail (%@gmail.com) or verify if customer names start with ‘A’.
Configuration Fields
Rule-Specific Configuration
| Field Name | Description | Required | Data Type / Options |
|---|
| SQL Like Pattern | SQL LIKE pattern to apply for value matching (use % and _) | Yes | String |
Success Criteria Configuration
This section defines how the rule’s outcome is measured against expected thresholds.
| Field Name | Description | Required | Options / Format |
|---|
| Operator | Comparison operation for the result count | Yes | GreaterThan, LessThan, EqualTo, Between |
| Threshold Value | Value for comparison (for GreaterThan, LessThan, EqualTo) | Conditional | Number |
| Threshold Min | Minimum value (for Between operator) | Conditional | Number |
| Threshold Max | Maximum value (for Between operator) | Conditional | Number |
| Is Percentage | Whether the threshold represents a percentage or an absolute count | No | true / false (default: false) |
| Allow Nulls | Whether null values are considered valid | No | true / false (default: false) |
| Check For Match | When false, validates for negation of the condition | No | true / false (default: true) |
Sample Configurations
Example 1: Email Domain Check
| Configuration Field | Value |
|---|
| Column | Email |
| SQL Like Pattern | %@gmail.com |
| Operator | GreaterThan |
| Threshold Value | 50 |
| Is Percentage | true |
| Allow Nulls | false |
| Check For Match | true |
Explanation:
Validates that more than 50% of emails belong to Gmail.
Example 2: Name Prefix Check
| Configuration Field | Value |
|---|
| Column | Name |
| SQL Like Pattern | A% |
| Operator | GreaterThan |
| Threshold Value | 0 |
| Is Percentage | false |
| Allow Nulls | false |
| Check For Match | true |
Explanation:
Validates that at least one name starts with ‘A’.
Sample Output
| Column Name | Rule Name | Success Count | Failure Count | Null Count | Within Threshold |
|---|
| Email | SQL Pattern Matching | 3 | 2 | 0 | Yes |
| Name | SQL Pattern Matching | 1 | 4 | 0 | Yes |
