aminghadersohi commented on PR #45598: URL: https://github.com/apache/superset/pull/45598#issuecomment-6333840535
Manual check on a local docker-compose-light env (examples dataset `cleaned_sales_data`, counts compared with SQL on the same database), head 37e5c78a8617b5ca52b73e172ba12673f03f2e93: - Cross-filters from the temporal ring now return exactly the bucket: raw 2003-11-14 → 38, Oct 2003 → 158, Nov 2004 → 301, Year 2004 → 1345, temporal outer ring Nov 2003 → 296, dashboard Time column + Time grain = Month on an ungrained chart → 158. On master every target failed with `invalid input syntax for type timestamp`. - Categorical rings: same filters and totals as master (967, 1384). Same results for a Gamma user with dataset access and an embedded guest. - Month labels include the year (Oct 2003 / Oct 2004); Drill by from an ungrained segment returns 2 / 19 / 17 rows (master: HTTP 400). - Drill to detail on a grained segment returned every row of the chart's time range (2823, or 2345 under a dashboard range): samples queries replace `TEMPORAL_RANGE` values with `time_range`. Commit 37e5c78a8617b5ca52b73e172ba12673f03f2e93 matches the bucket label with its grain instead, as Timeseries does; test `drill to detail matches a grained bucket by its label and grain` fails before it and passes after; live result 158. - Still open, not changed here: a grained Drill by has the same `time_range` problem (2823 / 2345 instead of 158). Timeseries x-axis Drill by on master does the same, so it needs a fix in the shared drill-by path. Also shared with Timeseries: a cross-filter on a bucket only partly inside the dashboard time range filters targets to the whole bucket. -- 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]
