[ 
https://issues.apache.org/jira/browse/OAK-4652?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16968136#comment-16968136
 ] 

Julian Reschke commented on OAK-4652:
-------------------------------------

trunk: (1.5.8) [r1755295|http://svn.apache.org/r1755295]

> indexName logged in QueryImpl is wrong in case of multiple indexes satisfying 
> the query
> ---------------------------------------------------------------------------------------
>
>                 Key: OAK-4652
>                 URL: https://issues.apache.org/jira/browse/OAK-4652
>             Project: Jackrabbit Oak
>          Issue Type: Bug
>          Components: query
>            Reporter: Chetan Mehrotra
>            Assignee: Chetan Mehrotra
>            Priority: Minor
>              Labels: candidate_oak_1_2, candidate_oak_1_4
>             Fix For: 1.5.8, 1.6.0
>
>
> QueryImpl currently computes the indexName like
> {code}
>  for (IndexPlan p : ipList) {
>                     // TODO limit is after all conditions
>                     long entryCount = Math.min(maxEntryCount, 
> p.getEstimatedEntryCount());
>                     double c = p.getCostPerExecution() + entryCount * 
> p.getCostPerEntry();
>                     if (c < cost) {
>                         cost = c;
>                         if (p.getPlanName() != null) {
>                             indexName += "[" + p.getPlanName() + "]";
>                         }
>                         indexPlan = p;
>                     }
>                 }
> {code}
> If there are multiple index plans then indexName contains multiple plan names 
> which causes confusion
> {noformat}
>    cost for 
> lucene-property[/oak:index/keymatchKeywords][/oak:index/newImport] is 786.0
> {noformat}
> Instead of concatenating the indexName should be generated outside



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to