Ivan Andika created HDDS-9338:
---------------------------------
Summary: Pagination framework in Recon
Key: HDDS-9338
URL: https://issues.apache.org/jira/browse/HDDS-9338
Project: Apache Ozone
Issue Type: Improvement
Reporter: Ivan Andika
>From [https://github.com/apache/ozone/pull/4585:]
"The pagination implementation can be implemented in different ticket (due to
the large change in this patch), but my current idea is to write a task to get
all the volumes / buckets from RocksDB tables, copying them to the separate
derby tables for volumes and buckets, and finally wrapping it in JPA
(pagination supported natively) / JOOQ (I'm not familiar with it, might need to
write sql for pagination). In that way, we can achieve the expected pagination
capabilities easily using known SQL methods (e.g. OFFSET and LIMIT) while still
retaining the ability to sort and search."
Currently Recon uses {{prevKey}} and {{limit}} to implement the pagination.
However, my concern is that the Recon pagination framework is quite different
from what is expected in the UI (antd table). For example, the UI sort by
attribute mechanism (e.g. bucket usedBytes) will not work since it's done in
memory and it's limited by the data returned by the backend.
The UI index-based pagination framework expected in the UI is more akin to JPA
framework's {{Pageable}} (and {{{}Sortable{}}}), but integrating this might
involve a very large change on the current Recon API since RocksDB does not
currently have some of the capabilities.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
