Ngone51 commented on issue #28177: [SPARK-31407][SQL][TEST] Fix hive/SQLQuerySuite.derived from Hive query file: drop_database_removes_partition_dirs.q URL: https://github.com/apache/spark/pull/28177#issuecomment-611944868 > Why don't we add sql("SELECT count(*) FROM src").queryExecution.analyzed at the very beginning of the test so it can initialize the table via TestHiveQueryExecution.analyzed in default database? So what about others tables? Initialize `src` table only just looks a bit weird to me. If we want to create tables before any tests we can simply create in `TestHiveQueryExecution` instead of current way? Another alternative way is to do something like below in `TestHiveQueryExecution` if it pass all tests: ``` try { // set currentDB to default loadTestTable(some_table) } finally { // restore original currentDB } ``` WDYT?
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
