Details
-
Improvement
-
Status: Open
-
Major
-
Resolution: Unresolved
-
2.5.2
-
None
-
None
Description
For storing large binary contents at times it is useful to store the binary directly to the datastore backend. For example when using S3 backed DataStore implementation it would be useful to directly upload the binary to the S3 and pass on the reference as part of JCR binary content.
As of now, api requires content's complete inputstream to create/save binary property in repository. The requirement is to provide binary implemenation which uses content reference and ensure that JR handles it in a mode similar to current support for BlobInDataStore
Use case:
1. Custom deployment in cloud using Amazon S3 backed data store
2. User leverages S3 capablity like multi-thread, multi-part upload [1] and uploads content to S3.
3. That content is then to be stored/referred in jcr:binary property in repository
[1] http://docs.amazonwebservices.com/AmazonS3/latest/dev/mpuoverview.html
Attachments
Issue Links
- relates to
-
JCR-3534 Efficient copying of binaries across repositories with the same data store
- Closed