[jira] [Commented] (OAK-4381) Query engine: decouple nodetype lookup

["Chetan Mehrotra (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/OAK-4381?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15293222#comment-15293222 ] 

Chetan Mehrotra commented on OAK-4381:
--------------------------------------

Thanks for taking this up! API looks fine. 

Can we route access to  {{NodeTypeInfoProvider}} via {{ExecutionContext}}. In oakutil [1] I provide a custom implementation of ExecutionContext which is then passed to QueryEngineImpl. If {{NodeTypeInfoProvider}}is obtained from that then I would be able to provide my own. Then I would not need to duplicate logic of {{QueryEngineImpl}}

[1] https://github.com/chetanmeh/oakutils/blob/master/src/main/java/com/chetanmeh/oak/index/config/generator/IndexConfigGenerator.java#L47

> Query engine: decouple nodetype lookup
> --------------------------------------
>
>                 Key: OAK-4381
>                 URL: https://issues.apache.org/jira/browse/OAK-4381
>             Project: Jackrabbit Oak
>          Issue Type: Improvement
>          Components: query
>            Reporter: Thomas Mueller
>            Assignee: Thomas Mueller
>             Fix For: 1.5.3
>
>         Attachments: OAK-4381.patch
>
>
> Currently, the query engine directly reads nodetype information from the relevant nodes from /jcr:system/jcr:nodeTypes. This strongly couples an implementation detail (the NodeState API) with the query engine.
> Instead, nodetype information should be encapsulated. That way, the parser can be used even if no repository with nodes states is available.



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