[jira] [Commented] (SPARK-15563) Packages specification

["Sean Owen (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-15563?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15302259#comment-15302259 ] 

Sean Owen commented on SPARK-15563:
-----------------------------------

I think it should go on the spark-packages site, but none of that doc source is part of this project. You should use the feedback link on that site to suggest this. Although having a standard build is a good thing, I don't know if all packages are even JVM-based, or if it's necessary to enforce this, as the site asserts nothing about the packages other than that someone submitted them.

> Packages specification
> ----------------------
>
>                 Key: SPARK-15563
>                 URL: https://issues.apache.org/jira/browse/SPARK-15563
>             Project: Spark
>          Issue Type: Improvement
>          Components: Documentation
>            Reporter: jay vyas
>
> PROBLEM
> Spark packages should be extensible as a mechanism for individual application developers to build productionable blueprints, and should be consumable/rebuildable/verifiable by robots.  Otherwise, there is a danger that a "spark package" might just be a link to a github repo without any useful code in it. 
> SOLUTION
> The spark-packages guidelines should enforce that either a `build.sbt`, `build.gradle`, `pom.xml`, or a `Makefile` present as part of the submission process.   Although this doesnt gaurantee one step, automatic builds, it is a pretty good measure of a projects status as a reproducible, maintainable artifact for the community.  
> I assume the spark-packages are a community maintained standard so the ASF Spark project is a good place to put this into the documentation.



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

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org