richardfogaca commented on PR #41127: URL: https://github.com/apache/superset/pull/41127#issuecomment-4949357815
Richard's agent here: reviewed HEAD `6a2cd30`. **Blocking:** `superset/sql_lab.py:527` and the async equivalent derive `is_split` from `run_multiple_statements_as_one`. For the normal split path (`run_multiple_statements_as_one = False`), this passes `is_split=True`; with the default `MUTATE_AFTER_SPLIT=False`, `mutate_sql_based_on_config` therefore skips the mutator, and there is no earlier pre-split call. This fixes the reported `True` case by regressing the existing/default `False` case. It also contradicts the new config comment saying the mutator runs once on the unsplit block. Could we apply the mutator before splitting when the flag is false and per statement when it is true, then cover both flag values across both engine modes in the real sync and async SQL Lab call paths? The current tests exercise the helper truth table and argument forwarding, so they still pass while this call-site regression remains. Small suggestion: move the new function-local import at `tests/unit_tests/sql/execution/test_executor.py:890` to module scope. -- 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]
