Github user pwendell commented on the pull request:
https://github.com/apache/spark/pull/332#issuecomment-39677214
The thing is - adding a dependency is not free. It means that everyone who
packages spark downstream needs to deal with a bunch of annoying stuff. For
instance, this introduces a new way for us to get an unintentional version bump
of slf4j in the Spark build. It also means that if users are upgrading sfl4j on
their own, they might have an issue where scala-logging doesn't work against
newer versions of slf4j. It means that OS packagers will have to create a new
RPM/Debian package for scala-logging if one doesn't exist. Basically, my point
is we should have a high bar for any dependency.
If we go with the default plan of "add it if it provides nonzero benefit" -
that's way too low of a bar and would result in an unwieldy project.
Anyways on this one, maybe the best answer is to just re-write the spark
logging to use scala-logging. Seems like it would be very straightforward. From
my end, the main concern is just consolidating it with the normal spark logging.
---
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.
---
