Axis Label Sorting and Aggregation

[madankgit]

Bug description

When using a date column (date_value) with Time Grain set to Day and applying time format %a (weekday abbreviation), the X-axis displays repeating weekday labels (Mon–Sun) instead of a single ordered weekday sequence.

Each individual date is correctly treated as a unique data point; however, the formatted axis labels repeat because %a controls display formatting only and does not group or aggregate dates by weekday.

Steps to Reproduce

1. Use a dataset containing a date column (date_value) with data spanning multiple weeks

2. Create a time-series chart

3. Configure:

   • Time column: date_value
   • Time Grain: Day
   • X-axis time format: %a

4. Observe the X-axis labels

Expected Behavior

• X-axis should display one ordered set of weekday labels (Mon–Sun) when %a is used
  OR
• Superset should provide an option to group by weekday when formatting by weekday name

Actual Behavior

• Weekday labels (Mon–Sun) are repeated for each week present in the dataset
• Results in duplicate and confusing X-axis labels

Impact

• Weekday-based analysis becomes unclear

• Clean visualization of weekly patterns is not possible

• Requires dataset-level workarounds using calculated columns

• Current-week or specific-week analysis is only possible via custom logic
  
  
  

Screenshots/recordings

No response

Superset version

master / latest-dev

Python version

3.9

Node version

16

Browser

Chrome

Additional context

No response

Checklist

• I have searched Superset docs and Slack and didn't find a solution to my problem.
• I have searched the GitHub issue tracker and didn't find a similar bug report.
• I have checked Superset's logs for errors and if I found a relevant Python stacktrace, I included it here as text in the "additional context" section.

[dosubot]

Hi @madankgit! I'm Dosu and I’m helping the superset team.

This happens because setting the X-axis time format to %a only changes how each date is displayed—it doesn't change how the data is grouped or aggregated. With time grain set to "Day," each date remains a unique data point, so you get repeated weekday labels for each week in your data. Superset doesn't currently have a built-in option to group or aggregate by weekday (e.g., all Mondays together) in the chart UI; grouping by weekday requires explicit aggregation logic in your dataset or query, not just display formatting [source].

To group by weekday, you can add a calculated column to your dataset (or use custom SQL in Explore) that extracts the weekday name or number from your date column, then use this new column as your group by field in the chart. For example, in SQL: EXTRACT(DOW FROM date_value) or TO_CHAR(date_value, 'Day') depending on your database. This will let you aggregate all Mondays, Tuesdays, etc., together and display a single ordered set of weekday labels [source].

There are ongoing discussions about making dynamic group by easier in the future, but as of now, this workaround is the recommended approach [source]. For best results, use both the weekday number (for sorting) and the name (for labeling) to ensure correct order on the X-axis.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

   

[madankgit]

Yes, the same issue occurs in these cases as well.

1. Month name (Jan–Dec)
When formatting the X-axis as Month name (Jan–Dec) with time grain set to Month, Superset still groups by month-year, so month labels repeat across years instead of aggregating all Januaries, Februaries, etc.

2. Hour of day (12 AM–11 PM)
Similarly, formatting the X-axis as Hour of day (12 AM–11 PM) with time grain set to Hour only changes the display. Each date-hour remains a separate data point, resulting in repeated hour labels unless day/hour is explicitly extracted and grouped in the dataset or SQL.