Cleanup Matches Task
Learn about the Cleanup Matches Task, which removes specified match data from the MATCHES table for specified match groups.
The Cleanup Matches task is a background job which is used to clean up the MATCHES table of the information related to specified match groups.
The API runs the matches cleanup job which cleans up the
MATCHES table of the information about potential matches, auto not matches, and grouping matches related to specified match groups. Tip:
- During implementation, the match groups may be suboptimal. If a tenant has a large volume of potential match pairs and the match group logic has been updated, it's recommended to drop the outdated potential matches using this job. This prevents the matching process from needing to remove them during recalculation. The job handles this cleanup much faster.
- This job is relatively lightweight and performs well when run with eight or more parts.
- The job removes potential matches from the matching storage, but it does not update the potential matches index. To update match counters, run the Potential Matches Reindex Task afterward.
- You can also use this job to drop data related to match groups of type
automatic_grouping. It deletes theassociationsandassociation linkswhich are internal objects used to manage grouped records. However, it does not remove the entities that represent the groups themselves. Use the Delete Entities Task to remove those entities.
To start this task, you must have Execute privileges.
Request
POST {{environment}}/reltio/api/{{tenantId}}/cleanupMatchesParameters
| Name | Required | Type | Description | Example |
|---|---|---|---|---|
entityType | Yes | String | Entity type ID. | Individual |
rules | yes | List [String] | Defines a comma-separated set of match group URIs to delete matches associated with them. | ["configuration/entityTypes/HCP/matchGroups/MatchByMiddleName", "configuration/entityTypes/HCP/matchGroups/MatchByMiddleNameAndLastName"] |
distributed | No | Boolean | The default value is false. If set to true, the job is run in distributed mode. | true |
taskPartsCount | No | Integer | The parameter specifies how many job parts are requested for the distributed mode. | 5 |
Request Example
{
"entityType": "Individual",
"rules": [
"configuration/entityTypes/Individual/matchGroups/SuspectMatchRule",
"configuration/entityTypes/Individual/matchGroups/GroupingMatchRule"
]
}
Response
Response has information about a created job. A job will have the following counters:
numberOfProcessedObjectstotalNumberOfDeletedMatches, which include:totalNumberOfDeletedMatchestotalNumberOfDeletedAssociationRecordstotalNumberOfDeletedAssociationLinks