EnxDev commented on PR #43938:
URL: https://github.com/apache/superset/pull/43938#issuecomment-5784865979

   Read through this one properly — it's a lot of surface area, but the core of 
it holds up. The locale problem that dominated the earlier rounds does look 
genuinely resolved rather than papered over: stored group keys are the literal 
`Main`, the chart keys are `t('Main')`, and `parseTimeComparisonSlot` / 
`resolveTimeComparisonSlotKeys` bridge the two in both directions. 
`isHeaderGroupsTimeComparisonEnabled` now matches the plugins' 
`isUsingTimeComparison` exactly (`time_compare` + aggregate + `Values`), 
`canSaveHeaderGroup` recurses into children, and `expandGroupColumnKey` 
short-circuits on an exact match before trying the `%` fallback, so the three 
things Copilot raised there are actually fixed in the diff and not just 
claimed. Nice touch stashing `allow_rearrange_columns` via the `visibility` 
gate instead of leaving it live to fight the group ordering.
   
   Six notes inline. The two I'd want addressed before merge are the prune 
effect in `HeaderGroupsControl` — it dispatches a non-`programmatic` `onChange` 
from mount, so opening a saved chart can flag unsaved changes the user never 
made, which is exactly what the `{ programmatic: true }` on your 
`ControlPanelsContainer` effect avoids — and `getClaimedColumnKeys` claiming 
`%${identifier}`, which locks a percent metric out of every other group as soon 
as its base metric is grouped. The rest are smaller: the auto-group label going 
stale when a metric is renamed, edit mode silently reverting an invalid edit 
with no feedback, a hardcoded `3px` radius, and a design question about 
table-specific logic landing in `ControlPanelsContainer`.
   
   Two things outside the diff worth a thought. This is a user-facing feature 
with no `docs/` update and no feature flag — is that deliberate? And the 
Testing Instructions section of the description is empty, which makes it harder 
for anyone to reproduce the time-comparison interaction by hand. Test coverage 
itself is solid; the header-group utils in particular are exercised well.
   
   <!-- enxdev-review-human:ba72da1 -->
   


-- 
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]

Reply via email to