carloea2 opened a new issue, #8249: URL: https://github.com/apache/texera/issues/8249
### Current behavior Python exposes byte values as unsigned integers, but Java byte arrays hash signed byte values. Binary tuple fields containing bytes above `0x7f` therefore produce different hashes in the Python and JVM engines, which can route the same hash-shuffle key to different workers. Before: the `0xff` key hashes to 317 in Python and 61 with Java semantics After: both engines hash the key to 61 ### Steps to reproduce On `50321e403c82df299a13deb50a7f9849dd93bdba`, hash a tuple with one binary field containing byte `0xff`. Observed values: ```text python_tuple_hash=317 java_compatible_tuple_hash=61 ``` ### Expected behavior Python binary tuple hashing should convert bytes to signed Java values before applying the array hash so shuffle routing matches the JVM. -- 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]
