Github user pwendell commented on the pull request:
https://github.com/apache/spark/pull/2673#issuecomment-58139100
Hey @vanzin this is something we need to support cross building in 2.10 and
2.11, but it's also a broadly useful thing that is lacking in the maven build
ecosystem (SBT has this feature out of the box).
The reason why it's needed is that we need to publish different dependency
trees for every artifact, say `spark-core` for both 2.10 and 2.11. The maven
build tool isn't good at distinguishing the published artifacts from the build
itself, so this will allow us to publish only the "effective" dependency graph
for a particular profile that is activated during publishing.
The reason why it's useful more broadly is that we have a lot of profiles
in Spark's build that get published with our poms, even though these are not
usable by projects that depend on Spark (profiles are not a concept that works
transitively across projects - or at least many build tools such as sbt do not
respect profiles from other projects). They just make our poms more confusing
looking for people. What we really want to publish is a single set of
dependencies for each Spark artifact.
---
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]
