uschindler commented on PR #13825: URL: https://github.com/apache/lucene/pull/13825#issuecomment-2380624362
> Thank you @rmuir, Maybe we can move `readGroupVInts()` into `GroupVIntUtil` to achieve the same goal? I raised a PR #13830 This is much better. Let's keep only the inner loop part in DataInput and replace the protected method by a public method called by the static utility reading the whole think. Anyways: I was a bit against the whole change from beginning because I was not happy to add new methods to DataInput just for a silly optimization. I am also not happy about the horrible bottom-up code just to get MemorySegment access working. IMHO, we should move minimum Java version of Lucene main branch (and 10.x) to Java 22 and introduce MemorySegment as a first class citizen and remove all abstractions from DataInput. Just make DataInput read from/to MemorySegment (NIOFSDir) or let it provide "views" (MMap) on the internal buffers. Let's move all our DataInput encoding/decoding logic to work on top of MemorySegment. That's all I have to say. I don't care about those arguments "wuhaaah, but Java 21 is LTS !!!!1!1111!!!!!!!!!!". Don't care, use Java 22 because this is what we need to be ready for future. Uwe -- 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: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
