Github user JoshRosen commented on the pull request:
https://github.com/apache/spark/pull/6515#issuecomment-115018376
I've thought about this a bit and I'm not sure that ErrorProne offers us a
huge win unless it's run as part of every build. I think the advantage of
ErrorProne over FindBugs is that it's fast, returns few false-positives, and
integrates into everyday builds. I think that we lose a lot of these advantages
if we guard its use behind a separate Jenkins trigger phrase or build profile /
flag; if we have to manually trigger it, then I feel that we might as well run
FindBugs instead.
I don't think that we can necessarily justify adding 10 minutes of build
time to the pull request builder in order to run these tests, since the
cost/benefit ratio there seems low. _However_, I'd be interested to see if
there's a way to compile _just_ the Java sources with this, but that might not
be possible given the Scala <-> Java interdependencies in our code base.
In light of all of this, I think that we might want to put this on hold for
now and revisit later once we have more flexible Jenkins triggering logic or
the ability to run the pull request builder tests using Maven.
Thanks for working on this and sorry that we're probably not going to be
able to pull it in.
---
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]
