bbeaudreault edited a comment on pull request #4304: URL: https://github.com/apache/hbase/pull/4304#issuecomment-1086308405
This is actually tougher than it seems. While this approach "works" in that the above goals are met, I discovered a side-effect which is that the new jars (both byo-hadoop and mapreduce) had each almost doubled in size. This is because while my new "provided-dependencies" modules succeed in ensuring the appropriate top-level dependencies in the resulting dependency-reduced-pom, they also cause the shade plugin to start shading all of the transitive dependencies of hadoop-hdfs, etc. I was able to mitigate this by adding a million exclusions, but it seems like this is not what we want. I'm trying to think of other approaches here but so far am coming up blank. It almost feels like we should move to a model if using `includes` to specify what exactly we want in the jar, rather than `excludes`. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
