[
https://issues.apache.org/jira/browse/SPARK-10183?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Amos updated SPARK-10183:
-------------------------
Priority: Minor (was: Major)
> Expose the SparkR backend api
> -----------------------------
>
> Key: SPARK-10183
> URL: https://issues.apache.org/jira/browse/SPARK-10183
> Project: Spark
> Issue Type: Improvement
> Reporter: Amos
> Priority: Minor
>
> The Backend class is currently scoped to the api.r package. I'm accessing it,
> for the Zeppelin project, so I can start SparkR against an already-running
> spark context. To do this I've had to create a helper class withing api.r. It
> would be better if the backend were exposed. It isn't a tremendous amount of
> functionality - create a backend, start it, stop it.
> (If we want to be really clever, it could also be passed a spark context and
> make that available to R clients, facilitate passing rdd's back and forth,
> etc. I'll be pushing code that does some of that to Zeppelin in a day or two
> if that helps.)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
