You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@drill.apache.org by "Steven Phillips (JIRA)" <ji...@apache.org> on 2014/10/21 23:33:33 UTC

[jira] [Commented] (DRILL-1382) Fast Schema Return

    [ https://issues.apache.org/jira/browse/DRILL-1382?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14179154#comment-14179154 ] 

Steven Phillips commented on DRILL-1382:
----------------------------------------

The basic design is as such:

buildSchema() method is added to RecordBatch interface. This method will call the same on incoming batches, and use upstream schema info to populate outgoing VectorContainer, making schema info available to downstream operator. 

the buildSchema() method of RootExec operators will also be responsible for sending empty schema batches downstream.

Fragment executor will call buildSchema() on the RootExec before calling next();

> Fast Schema Return
> ------------------
>
>                 Key: DRILL-1382
>                 URL: https://issues.apache.org/jira/browse/DRILL-1382
>             Project: Apache Drill
>          Issue Type: Improvement
>          Components: Execution - Flow
>            Reporter: Jacques Nadeau
>            Assignee: Steven Phillips
>             Fix For: 0.6.0
>
>
> Right now, the first schema for a query set is not returned until all blocking operators complete.  In many cases, it would be useful for an empty record set to be returned prior to completing blocking operators so that tools can understand schema sooner.  



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