Process Rebuilt Match Task
Information about when and how to run this task.
The Spark match job must write information about objects that are processed by the platform matching process. This includes information about automatic matches, matches from relevance_based match groups, and matches from custom match groups.
A new task
ProcessRebuiltMatchTask is introduced that provides: - Out-of-the-box support for all match group types
- The ability to issue events
- The ability to handle rejects properly
RebuildMatchTableV2Task task, groups them by rule type (automatic,
suspect, relevance), and invokes the relevant handlers. Request
Administrator role is required.
POST /api/{tenantId}/processRebuiltMatches| Name | Type | Default |
|---|---|---|
distributed |
Boolean | False |
taskPartsCount |
Integer | 1 |
The ProcessRebuiltMatchTask task can be run automatically after the
RebuildMatchTableV2Task task if the tenant's physical configuration
contains matchJob: AGGREGATING_BY_DOCUMENT_MATCH as follows:
{
"matchingConfiguration": {
"rebuildMatchTableV2": {
"matchJob": "AGGREGATING_BY_DOCUMENT_MATCH",
"writeMatchesBatchSize": 150,
"overcollisionedTokenThreshold": 300
}
}
}