[
https://issues.apache.org/jira/browse/SLING-3836?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14092718#comment-14092718
]
Marius Petria commented on SLING-3836:
--------------------------------------
There are two options to implement this. At the level of replication API or as
a configuration of the package exporter.
- API based: Add a property (isDeep) to ReplicationRequest that lets the agent
know what the intention is.
- Configuration based: An agent can be configured to replicate all requests
deep or shallow. A client will have to use two separate agents for replicating
deep or shallow.
It is worth thinking how such options should be treated in general. We might
need to add options for synchronous replication, or even more complex tree
replication (always replicating jcr:content for pages).
> Allow deep vs shallow replication options
> -----------------------------------------
>
> Key: SLING-3836
> URL: https://issues.apache.org/jira/browse/SLING-3836
> Project: Sling
> Issue Type: Improvement
> Components: Replication
> Reporter: Marius Petria
> Labels: replication
>
> Users of replication should have a way to choose whether to replicate a node
> - deep: including all its subtree
> - shallow: including just he node properties
> Typically an user editing content visually will trigger a replication of only
> the edited content (shallow).
> However an user importing a lot of content will trigger a deep replication
> after all content is imported.
> Shallow replication is more targeted and produces the least changes in the
> system and it is probably best fitted to be the default replication option.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
