srielau commented on PR #58317: URL: https://github.com/apache/spark/pull/58317#issuecomment-5821105920
Addressed the latest review in d161b01b76c: - Direct V2 caches named with a short identifier are restored on rename by matching resolved table identity, including analyzer-generated CHAR/VARCHAR projections. - V2 rename restoration clones the session instead of mutating the caller SQLConf. - Descriptor helper Scaladoc now covers both call forms. - V1 insert and save recache tests bind PreserveNative and SparkStandard. - Hive ORC cache identity is covered for converted and native tables. - `CharVarcharScanMode.None` docs now say no mode was bound, including relations with no CHAR/VARCHAR columns. Unified CacheManager lifecycle work (one descriptor for V1/V2/Hive mutation, refresh, and rename, plus isolated restoration as a general contract) remains in [SPARK-59751](https://issues.apache.org/jira/browse/SPARK-59751). The Kafka CI failure on the previous head (`KafkaMicroBatchV2SourceWithConsumerSuite` timing out in `waitUntilMetadataIsPropagated`) is unrelated: this PR does not touch Kafka sources. The new push retriggers CI. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
