HyukjinKwon edited a comment on issue #27307: [SPARK-30572][SPARK-30534][BUILD][FOLLOW-UP] Add a fallback repository as GCS URL: https://github.com/apache/spark/pull/27307#issuecomment-576720395 @tgravescs, just for clarification, this change is needed to address all of three. So, even if `sbt-pom-reader` fixes the issue, this change here will still be needed to address other two bullet points in the PR description. For the sbt issue, I am a bit hesitant to explain because I am not super confident of my answer so just take it as a speculation. _I think_: 1. `sbt-pom-reader` is loaded as a SBT plugin. 2. It tries to convert POM file to SBT project structure 3. During this process, it overwrites the default `central` maven resolver at [here](https://github.com/JoshRosen/sbt-pom-reader/blob/v1.0.0-spark/src/main/scala/com/typesafe/sbt/pom/MavenPomResolver.scala#L32), which is `http`. 4. The conversion is done. 5. Now, we use SBT to resolve the dependencies by the resolvers defined [here](https://github.com/apache/spark/blob/master/project/SparkBuild.scala#L227), which is [`https://... `](https://www.scala-sbt.org/0.13/docs/Resolvers.html). Before this fix, at 3., `central` is `http` and there's no other repos to fall back. So it fails. After this fix, at 3., `google-manve-central` becomes a fall back so it can handle during 3. And yes, currently, I think it's an issue in `sbt-pom-reader` too.
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
