[jira] [Commented] (OAK-4162) Nodetype index is used even if nodetype is not indexed

2016-05-25 Thread Thomas Mueller (JIRA) 

[ 
https://issues.apache.org/jira/browse/OAK-4162?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15300068#comment-15300068
 ] 

Thomas Mueller commented on OAK-4162:
-

I found out the problem was slightly different: the nodetype (acme:Page) was 
not indexed directly, but a mixin was indexed (sling:VanityPath I think), and 
the nodetype acme:Page had sling:VanityPath as a mixin. So, indirectly the 
nodetype _was_ indexed. So not a bug.

> Nodetype index is used even if nodetype is not indexed
> --
>
> Key: OAK-4162
> URL: https://issues.apache.org/jira/browse/OAK-4162
> Project: Jackrabbit Oak
>  Issue Type: Bug
>Reporter: Thomas Mueller
>Assignee: Thomas Mueller
>
> If the nodetype index _did_ in the past index a certain nodetype (for example 
> acme:Page), but that nodetype was removed from the index definition, then the 
> nodetype index is still used for queries with that nodetype, for example:
> {noformat}
> /jcr:root//element(*, acme:Page)
> {noformat}
> It looks like the index is used because there is a node 
> /oak:index/nodetype/:index/acme%3APage.
> A workaround is to reindex (re-build) the nodetype index.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

[jira] [Commented] (OAK-4162) Nodetype index is used even if nodetype is not indexed

2016-03-31 Thread Thomas Mueller (JIRA) 

[ 
https://issues.apache.org/jira/browse/OAK-4162?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15219932#comment-15219932
 ] 

Thomas Mueller commented on OAK-4162:
-

Hm, maybe what I wrote above is not actually true... Need to do more checks 
first.

> Nodetype index is used even if nodetype is not indexed
> --
>
> Key: OAK-4162
> URL: https://issues.apache.org/jira/browse/OAK-4162
> Project: Jackrabbit Oak
>  Issue Type: Bug
>Reporter: Thomas Mueller
>Assignee: Thomas Mueller
>
> If the nodetype index _did_ in the past index a certain nodetype (for example 
> acme:Page), but that nodetype was removed from the index definition, then the 
> nodetype index is still used for queries with that nodetype, for example:
> {noformat}
> /jcr:root//element(*, acme:Page)
> {noformat}
> It looks like the index is used because there is a node 
> /oak:index/nodetype/:index/acme%3APage.
> A workaround is to reindex (re-build) the nodetype index.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)