[
https://issues.apache.org/jira/browse/LENS-473?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14392612#comment-14392612
]
Hudson commented on LENS-473:
-----------------------------
FAILURE: Integrated in Lens-Commit #680 (See
[https://builds.apache.org/job/Lens-Commit/680/])
LENS-473 : Change the queryable field validation logic (jdhok: rev
019eb9498b54512e28e0e85a33a9afad814b1c84)
* lens-cube/src/main/java/org/apache/lens/cube/parse/FieldValidator.java
* lens-cube/src/test/java/org/apache/lens/cube/parse/TestQueryMetrics.java
* lens-cube/src/test/java/org/apache/lens/cube/parse/TestBaseCubeQueries.java
* lens-cube/src/main/java/org/apache/lens/cube/parse/CubeQueryRewriter.java
* lens-cube/src/test/java/org/apache/lens/cube/parse/TestQueryRewrite.java
* lens-cube/src/main/java/org/apache/lens/cube/parse/AliasReplacer.java
> Change the queryable field validation logic
> -------------------------------------------
>
> Key: LENS-473
> URL: https://issues.apache.org/jira/browse/LENS-473
> Project: Apache Lens
> Issue Type: Improvement
> Components: cube
> Affects Versions: 2.1
> Reporter: Amareshwari Sriramadasu
> Assignee: Jaideep Dhok
> Priority: Critical
> Fix For: 2.1
>
> Attachments: lens-473.patch
>
>
> Right now the field validation logic at
> https://github.com/apache/incubator-lens/blob/master/lens-cube/src/main/java/org/apache/lens/cube/parse/AliasReplacer.java#L103
> does the following :
> It gets all the dim-attributes queried and measures queried in the user query
> (after expressions are expanded), then it finds if there is single derived
> cube having all the attributes queried and any of measures queried. And makes
> sure all the measures queried are covered by attribute covering derived
> cubes. For chain attributes, it picks source columns of chains as the
> dim-attributes in the above validate.
> Instead it should do the following :
> * Do validation only when query has aggregates
> * Pick dim-attributes (or chain ref columns) only if they are not part of
> aggregate expression.
> * Above mentioned logic of attribute covering and measuring covering logic
> would be the same.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)