[
https://issues.apache.org/jira/browse/JCR-3750?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13938864#comment-13938864
]
Amit Jain commented on JCR-3750:
--------------------------------
Hi [~tmueller], [~chetanm],
My proposal is to add the following method in the DataStore and Backend API
{code}void deleteRecords(List<DataIdentifier> ids, long maxModifiedTime){code}.
Implementations will then handle the batch deletion in the best possible way.
Currently, DataStore does not have an API for deletion based on an identifier
but MultiDataStoreAware has it. I find it preferable to add it to the
DataStore, but if you think its better to add it to the MultiDataStoreAware,
then I'll use that.
> Add a batch delete method for the data stores.
> ----------------------------------------------
>
> Key: JCR-3750
> URL: https://issues.apache.org/jira/browse/JCR-3750
> Project: Jackrabbit Content Repository
> Issue Type: Bug
> Components: jackrabbit-data
> Affects Versions: 2.7.5
> Reporter: Amit Jain
>
> Currently, MutiDataStoreAware exposes a {code}deleteRecord(DataIdentifier
> identifier){code} method to delete a blob given an id.
> The proposal is to add a new delete method which can take in a batch of ids
> and take advantage of the underlying backends capability to delete in batches
> (if supported). Backends like S3 and DB can handle batch delete.
> This will help in delete performance when a large number of deletes are to
> happen for ex. during garbage collection.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
