Uh oh!
There was an error while loading. Please reload this page.
[SPARK-19572][SPARKR] Allow to disable hive in sparkR shell - #16907
[SPARK-19572][SPARKR] Allow to disable hive in sparkR shell#16907zjffdu wants to merge 3 commits into
Conversation
SparkQA
commented
Feb 13, 2017
Test build #72801 has started for PR 16907 at commit |
SparkQA
commented
Feb 13, 2017
Test build #72810 has finished for PR 16907 at commit
|
zjffdu
commented
Feb 13, 2017
@felixcheung Please help review. |
| } else { | ||
| if (enableHiveSupport) { | ||
| if (enableHiveSupport | ||
| && jsc.sc.conf.get(CATALOG_IMPLEMENTATION.key, "hive") == "hive") { |
There was a problem hiding this comment.
I don't think we should check for this for this message
I think we should have a message for hive support being turned off by conf. Otherwise it will be confusing for someone running sparkR.session(enableHiveSupport = TRUE) and didn't get it.
There was a problem hiding this comment.
you need to toLowerCase like here https://github.com/apache/spark/blob/master/repl/scala-2.11/src/main/scala/org/apache/spark/repl/Main.scala#L91
felixcheung
commented
Feb 13, 2017
We should discuss other uses outside of sparkR shell - shell.R is not intended to be an API for reuse outside of the sparkR shell - we have other assumptions in the code that could be easily broken that way. |
| } else { | ||
| if (enableHiveSupport) { | ||
| if (enableHiveSupport | ||
| && jsc.sc.conf.get(CATALOG_IMPLEMENTATION.key, "hive") == "hive") { |
There was a problem hiding this comment.
you need to toLowerCase like here https://github.com/apache/spark/blob/master/repl/scala-2.11/src/main/scala/org/apache/spark/repl/Main.scala#L91
zjffdu
commented
Feb 14, 2017
Address the comments. @felixcheung, correct, |
There was a problem hiding this comment.
I'd be specific to say instead of Hive is disabled via ${CATALOG_IMPLEMENTATION.key}
say ${CATALOG_IMPLEMENTATION.key} is not set to "hive";
SparkQA
commented
Feb 14, 2017
Test build #72842 has finished for PR 16907 at commit
|
SparkQA
commented
Feb 15, 2017
Test build #72915 has finished for PR 16907 at commit
|
zjffdu
commented
Feb 24, 2017
Seems a flaky test, let me trigger the build |
SparkQA
commented
Feb 24, 2017
Test build #73416 has finished for PR 16907 at commit
|
felixcheung
left a comment
There was a problem hiding this comment.
LGTM.
You want this in branch-2.1 and master, yes?
zjffdu
commented
Feb 25, 2017
Yeah, it would be nice to be merged into 2.1 as well. Thanks |
| && jsc.sc.conf.get(CATALOG_IMPLEMENTATION.key, "hive").toLowerCase == "hive") { | ||
| logWarning("SparkR: enableHiveSupport is requested for SparkSession but " + | ||
| "Spark is not built with Hive; falling back to without Hive support.") | ||
| s"Spark is not built with Hive or ${CATALOG_IMPLEMENTATION.key} is not set to 'hive', " + |
There was a problem hiding this comment.
actually, I notice this when reviewing before merge, I think this check is not matching with the message - in this case we already know CATALOG_IMPLEMENTATION == "hive"
There was a problem hiding this comment.
Good catch, I updated the if condition to match the message.
SparkQA
commented
Mar 1, 2017
Test build #73642 has finished for PR 16907 at commit
|
## What changes were proposed in this pull request? SPARK-15236 do this for scala shell, this ticket is for sparkR shell. This is not only for sparkR itself, but can also benefit downstream project like livy which use shell.R for its interactive session. For now, livy has no control of whether enable hive or not. ## How was this patch tested? Tested it manually, run `bin/sparkR --master local --conf spark.sql.catalogImplementation=in-memory` and verify hive is not enabled. Author: Jeff Zhang <zjffdu@apache.org> Closes#16907 from zjffdu/SPARK-19572. (cherry picked from commit 7315880) Signed-off-by: Felix Cheung <felixcheung@apache.org>
felixcheung
commented
Mar 1, 2017
merged to master and branch-2.1 |
What changes were proposed in this pull request?
SPARK-15236 do this for scala shell, this ticket is for sparkR shell. This is not only for sparkR itself, but can also benefit downstream project like livy which use shell.R for its interactive session. For now, livy has no control of whether enable hive or not.
How was this patch tested?
Tested it manually, run
bin/sparkR --master local --conf spark.sql.catalogImplementation=in-memoryand verify hive is not enabled.