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 "Chetan Mehrotra (JIRA)" <ji...@apache.org> on 2017/10/16 10:56:00 UTC

[jira] [Created] (OAK-6835) nodetype lucene index can create unusable nodes under index structures

Chetan Mehrotra created OAK-6835:
------------------------------------

             Summary: nodetype lucene index can create unusable nodes under index structures
                 Key: OAK-6835
                 URL: https://issues.apache.org/jira/browse/OAK-6835
             Project: Jackrabbit Oak
          Issue Type: Bug
          Components: property-index
            Reporter: Chetan Mehrotra
            Assignee: Vikas Saurabh
            Priority: Minor
             Fix For: 1.8


{{nodetype}} index is just a special property index definition with definition declaring to index {{jcr:primaryType}} and {{jcr:mixinTypes}}.
Since, it's just property index, we can specify {{declaringNodeTypes}} too to filter which type of nodes get indexed.

On query side, declaringNodeTypes are used to check whether the index can be used at all or not.

Now, for nodetype, if node being indexed passes declaringNodeType filter (either primary type or mixin matches), then all its mixins and primaryType gets indexed irrespective of whether declaringNodeType contains it or not.
This is perfectly correct behavior from property index point of view. But, in this regards, we should treat nodetype index to be special and index only those property values that are part of declaring node types.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)