peterdettman commented on pull request #242: URL: https://github.com/apache/commons-compress/pull/242#issuecomment-1026993646
I've only really looked at mainQSort3; we have a fork of an older version of this bzip2 code in the BouncyCastle codebase and I refactored the qsort code a bit, then noticed this discrepancy when referring back here. I don't think median selection affects correctness; it's only a matter of improving performance if you can choose better pivots on average. Even then I'm not sure it affects overall performance that much, although perhaps there is some subclass of inputs where it does, or some worst-case that is avoided. -- 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]
