Github user brkyvz commented on the pull request:
https://github.com/apache/spark/pull/7139#issuecomment-122025235
@sun-rui Thanks for the comments! This PR is basically aiming to support
complex Spark Packages, basically packages that have Scala / Java, Python,
and/or R code in them. Not just pure R code. Devs will basically write their
code in Scala, and provide R APIs. I believe @cafreeman is working on some
packages that have this structure.
1. We handle all Maven dependency related information through the
`--packages` flag. For example, a package @cafreeman is working on depends on
amplab/mlmatrix on the Scala side, and it has R APIs. We added the `--packages`
flag to handle situations like this. IMHO I'm against adding more Spark Submit
arguments, that would increase complexity. We have Jars that contain Python
code in addition to Scala, that is being handled through `--packages` right now.
2. I wanted to keep this PR simple, lay the ground work, and support those
cases in follow-up PRs. I can handle those cases in this PR, or follow up like
I intentionally planned. Depends on how big a PR you are willing to code review
:)
3. I was looking for that exact configuration, thanks! I'll add that.
4. I know, I believe we started working on these in parallel. I can reuse
the existing code there.
What do you think? How should I proceed on this PR?
---
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]
