aminghadersohi commented on PR #44695: URL: https://github.com/apache/superset/pull/44695#issuecomment-5903794401
Following up on the three additional suggestions in Bito run #f39aac, checked against head 45309e2e: - **Redundant second `fillna` pass:** gone. `df_to_sql` no longer calls `fillna`; it builds `values` once from `df.astype(object).where(df.notna(), None)` and maps missing cells to `""` (`superset/db_engine_specs/gsheets.py:645-651`, from ddfec16e01). - **Datetime/date branch order:** `to_json_value` tests `datetime` before `(date, time)`, so a `datetime` never falls into the date branch. `np.datetime64` goes through `pd.Timestamp` first, which covers nanosecond precision (`gsheets.py:69-84`, `:101-106`). Truncating to microseconds via `to_pydatetime()` is intended: `USER_ENTERED` doesn't parse finer precision. `test_upload_cell_types` pins the nanosecond case. - **Impersonation flag provenance:** no longer applicable. Since 4954f82e99, validation never reads `impersonate_user`. It always sends `subject=None` because service-account queries do the same (`gsheets.py:467-479`), and the create/edit regression covers all three flag states. -- 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]
