[PR] Split the Avro module into client, source, sink, serializer and spooldir [logging-flume-rpc]

["ppkarwasz (via GitHub)" <gi...@apache.org>]

ppkarwasz opened a new pull request, #15:
URL: https://github.com/apache/logging-flume-rpc/pull/15

   Same split for Avro: `flume-avro-client` only depends on `flume-client`, and the source and sink get their own artifacts under `avro/`. The parts unrelated to RPC also get their own artifacts: `flume-avro-serializer` for Avro container files and `flume-spooldir-source` for the spooling directory source and the `avro-client` command.
   
   The Avro protocol keeps its `org.apache.flume.source.avro` namespace: Avro 1.11 resolves enums by their full name, so renaming it breaks Flume 1.x clients talking to a 2.x agent, after the event is already accepted. This was checked with a 1.11.0 client and server in both directions. `AvroSource` therefore moves to `org.apache.flume.source.avro.server`.
   
   The classes generated by Avro and Thrift are excluded from SpotBugs, so the violation ceilings only cover maintained code.
   
   Stacked on the Thrift split (`feat/split-thrift`).
   
   🤖 Generated with [Claude Code](https://claude.com/claude-code)
   


-- 
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: notifications-unsubscribe@logging.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org