Github user jkleckner commented on a diff in the pull request:
https://github.com/apache/spark/pull/4780#discussion_r25440596
--- Diff: pom.xml ---
@@ -471,13 +471,6 @@
<groupId>com.clearspring.analytics</groupId>
<artifactId>stream</artifactId>
<version>2.7.0</version>
- <exclusions>
- <!-- Only HyperLogLogPlus is used, which doesn't depend on
fastutil -->
- <exclusion>
- <groupId>it.unimi.dsi</groupId>
- <artifactId>fastutil</artifactId>
--- End diff --
I'm very open to suggestions.
Note that parquet is already bringing in part of it via it's own assembly
jar renamed under parquet/.
I haven't worked with sbt-assembly rename, but is that a solution?
Perhaps we should all use a different implementation other than QDigest
like tdigest?
BTW, I'm unclear where discussion is best placed. Should it be here or
there: https://issues.apache.org/jira/browse/SPARK-6029
The more I look into this sort of ```dependency **ll```, the more I believe
it is intractable in the current way we (generally, not Spark specifically)
package things.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
