aminghadersohi commented on code in PR #41489:
URL: https://github.com/apache/superset/pull/41489#discussion_r3614133931
##########
superset/models/helpers.py:
##########
@@ -1625,6 +1625,56 @@ def _resolve(label: str | None) -> tuple[bool, str |
None]:
seen.add(label)
return tuple(labels)
+ def _offset_only_dttm_cols(
Review Comment:
NIT: no test exercises this with a nonzero `offset`/`time_shift` where a
column is *already* in `already_collected` (base-axis, `DTTM_ALIAS`, or a
`python_date_format`-declared raw column) and also native-datetime dtype in the
DataFrame. The existing dedup tests
(`test_normalize_df_dedups_column_in_granularity_and_columns`,
`test_normalize_df_normalizes_base_axis_temporal_columns`,
`test_normalize_df_normalizes_legacy_time_column`) all set `datasource.offset =
0`, so the `already_collected` skip here never runs under a truthy offset.
Worth a regression test asserting the primary time column shifts by exactly one
offset (not two) to pin the disjointness this fix relies on.
--
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]
