msyavuz commented on PR #43718: URL: https://github.com/apache/superset/pull/43718#issuecomment-5481093364
Both landed in 73ac365. **GROUPING SETS:** `drop_grouping_sets_rollups` keeps only rows whose markers are all 0 and drops the marker columns, called before `pivot_df` so the currency-context pivot sees the same frame. Test asserts the denominator is 30 rather than 60 and that no phantom row appears. **MIN/MAX:** each metric's reducer now drives both the totals and the denominators — it had to be both, since changing only the denominators makes the Total column divide a max by a sum and reinstates the 50% problem from the thread above. `aggfunc = "Sum"` stays for cell aggregation and the total's label. A MAX row of [6, 10] reads 60%/100%. One wrinkle worth noting: a total column carries a total label rather than a metric name, so resolving its reducer by its own label fell back to sum while its value had been rolled up with max — the grand total read 0.625 and the corner cell 1.3. Both now derive the reducer from the columns forming the denominator. Also added the `combineMetric` case and a cross-metric total assert. -- 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]
