You are viewing a plain text version of this content. The canonical link for it is here.
Posted to oak-issues@jackrabbit.apache.org by "Marcel Reutegger (JIRA)" <ji...@apache.org> on 2014/04/07 09:10:15 UTC

[jira] [Commented] (OAK-1678) document atomicity of DS.remove(collection, keys)

    [ https://issues.apache.org/jira/browse/OAK-1678?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13961671#comment-13961671 ] 

Marcel Reutegger commented on OAK-1678:
---------------------------------------

In case of an error, the implementation does not have to guarantee atomicity of the
overall call. Only operation on the document level must be atomic. I guess it would
be good to specify this general requirement on the DocumentStore class level.

I think it should be up to the implementation to decide whether to continue removing
documents when an error occurs. So, this is kind of best effort.

It is not an error if a document doesn't exist for a given key. The implementation
must simply ignore those keys.

> document atomicity of DS.remove(collection, keys)
> -------------------------------------------------
>
>                 Key: OAK-1678
>                 URL: https://issues.apache.org/jira/browse/OAK-1678
>             Project: Jackrabbit Oak
>          Issue Type: Task
>            Reporter: Julian Reschke
>              Labels: documentation
>
> [~mreutegg]
> I believe it's best effort (looking at the MongoDB impl), but it would be good to clarify.
> In particular, should the operation abort then one removal failed, or keep going? What's the expectation when a document doesn't exist?



--
This message was sent by Atlassian JIRA
(v6.2#6252)