Miuler commented on PR #23610:
URL: https://github.com/apache/beam/pull/23610#issuecomment-1373754159
> Oh hmm when I last commented I am not sure what I was thinking actually,
and I don't remember noticing the scala dependency. What does this mean for
Java users? The built artifact will have a dependency on a very specific
version of the scala runtime and libraries, right? I don't think Beam has
really done much work to manage different scala versions effectively. Though
our runners do also use scala...
The version used in most cases is 2.12, although there are still some 2.11,
which should be updated,
```
./sdks/java/io/azure-tablestorage/build.gradle:37:
implementation("org.scala-lang:scala-library:2.12.17")
./sdks/java/io/azure-cosmosdb/build.gradle:38:
implementation("org.scala-lang:scala-library:2.12.17")
./sdks/java/io/sparkreceiver/2/build.gradle:50: compileOnly
"org.scala-lang:scala-library:2.11.12"
./runners/spark/spark_runner.gradle:179: compileOnly
"org.scala-lang:scala-library:2.11.12"
./runners/spark/spark_runner.gradle:183: compileOnly
"org.scala-lang:scala-library:2.12.15"
./runners/spark/3/job-server/build.gradle:38: force
"org.scala-lang:scala-library:2.12.10"
./runners/samza/build.gradle:59: runtimeOnly
"org.scala-lang:scala-library:2.11.8"
```
But it would be nice to work on gradle, so you can cross-compile for 2.12,
2.13, and it would be transparent for the current, 3.x version not yet, because
neither spark nor flink support it yet.
--
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]
