Heatmap with "Normalize Across" x/y will fail when filters reduce the data to a single column/row.

[EUbaldiEC]

Bug description

When creating heatmaps (both tested in 6.0.0 and 6.1.0) with normalisation across X/Y (and posibly color normalisation), the plot will fail when only one column/row is left by applying dynamic filters.

Example is in the flights sample dataset, on a fresh install, putting AIRLINE as the X and CITY as Y (metric COUNT(*) or SUM(AIRLINE_DELAY) or anything else).

This might be related to the underlying eChart heatmap implementation (and the error thrown stems from pandas postprocessing rank function) but I am struggling to understand where to look in the codebase to check for issues.

Let me know should you need any additional information, thanks!

Screenshots/recordings

Example is in the flights sample dataset, on a fresh install, putting AIRLINE as the X and CITY as Y (metric COUNT(*) or SUM(AIRLINE_DELAY) or anything else).

It works as expected with the normalisation across X (above), but as soon as a filter reducing the data to a single column (like AIRLINE = 'AA'), a "Cannot set a DataFrame with multiple columns to the single column rank" error is shown.

The weird thing is that then the plot works with no issues if one normalises across rows/cols (Y/X) (the same works when normalising across heatmap):

To conclude, an unexptected behavior is also found when reducing both the rows and the columns to 1:

• If one normalises across X/Y we get an empty plot (should plot a single square with 100% in both cases)

• Instead, when normalising over the heatmap this is working as expected:

Superset version

6.0.0

Python version

3.10

Node version

18 or greater

Browser

Not applicable

Additional context

Here is the python stacktrace from container logs:

2026-06-03 12:27:08,053:WARNING:superset.views.error_handling:Exception
Traceback (most recent call last):
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/flask/app.py", line 917, in full_dispatch_request
    rv = self.dispatch_request()
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/flask/app.py", line 902, in dispatch_request
    return self.ensure_sync(self.view_functions[rule.endpoint])(**view_args)  # type: ignore[no-any-return]
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/flask_appbuilder/security/decorators.py", line 101, in wraps
    return f(self, *args, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/views/base_api.py", line 120, in wraps
    duration, response = time_function(f, self, *args, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/utils/core.py", line 1559, in time_function
    response = func(*args, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/utils/log.py", line 299, in wrapper
    value = f(*args, add_extra_log_payload=log, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/charts/data/api.py", line 289, in data
    return self._get_data_response(
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/utils/log.py", line 301, in wrapper
    value = f(*args, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/charts/data/api.py", line 496, in _get_data_response
    result = command.run(force_cached=force_cached)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/commands/chart/data/get_data_command.py", line 46, in run
    payload = self._query_context.get_payload(
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context.py", line 99, in get_payload
    return self._processor.get_payload(cache_query_context, force_cached)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context_processor.py", line 353, in get_payload
    query_results = [
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context_processor.py", line 354, in <listcomp>
    get_query_results(
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_actions.py", line 290, in get_query_results
    return result_func(query_context, query_obj, force_cached)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_actions.py", line 160, in _get_full
    payload = query_context.get_df_payload(query_obj, force_cached=force_cached)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context.py", line 129, in get_df_payload
    return self._processor.get_df_payload(
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context_processor.py", line 131, in get_df_payload
    query_result = self.get_query_result(query_obj)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context_processor.py", line 243, in get_query_result
    return self._qc_datasource.get_query_result(query_object)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/models/helpers.py", line 1343, in get_query_result
    df = query_object.exec_post_processing(df)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_object.py", line 542, in exec_post_processing
    df = getattr(pandas_postprocessing, operation)(df, **options)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/utils/pandas_postprocessing/rank.py", line 37, in rank
    df["rank"] = gb.apply(lambda x: x[metric].rank(pct=True))
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/pandas/core/frame.py", line 4081, in __setitem__
    self._set_item_frame_value(key, value)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/pandas/core/frame.py", line 4239, in _set_item_frame_value
    raise ValueError(
ValueError: Cannot set a DataFrame with multiple columns to the single column rank
2026-06-03 12:27:08,053:ERROR:superset.views.error_handling:Cannot set a DataFrame with multiple columns to the single column rank
Traceback (most recent call last):
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/flask/app.py", line 917, in full_dispatch_request
    rv = self.dispatch_request()
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/flask/app.py", line 902, in dispatch_request
    return self.ensure_sync(self.view_functions[rule.endpoint])(**view_args)  # type: ignore[no-any-return]
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/flask_appbuilder/security/decorators.py", line 101, in wraps
    return f(self, *args, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/views/base_api.py", line 120, in wraps
    duration, response = time_function(f, self, *args, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/utils/core.py", line 1559, in time_function
    response = func(*args, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/utils/log.py", line 299, in wrapper
    value = f(*args, add_extra_log_payload=log, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/charts/data/api.py", line 289, in data
    return self._get_data_response(
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/utils/log.py", line 301, in wrapper
    value = f(*args, **kwargs)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/charts/data/api.py", line 496, in _get_data_response
    result = command.run(force_cached=force_cached)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/commands/chart/data/get_data_command.py", line 46, in run
    payload = self._query_context.get_payload(
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context.py", line 99, in get_payload
    return self._processor.get_payload(cache_query_context, force_cached)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context_processor.py", line 353, in get_payload
    query_results = [
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context_processor.py", line 354, in <listcomp>
    get_query_results(
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_actions.py", line 290, in get_query_results
    return result_func(query_context, query_obj, force_cached)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_actions.py", line 160, in _get_full
    payload = query_context.get_df_payload(query_obj, force_cached=force_cached)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context.py", line 129, in get_df_payload
    return self._processor.get_df_payload(
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context_processor.py", line 131, in get_df_payload
    query_result = self.get_query_result(query_obj)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_context_processor.py", line 243, in get_query_result
    return self._qc_datasource.get_query_result(query_object)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/models/helpers.py", line 1343, in get_query_result
    df = query_object.exec_post_processing(df)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/common/query_object.py", line 542, in exec_post_processing
    df = getattr(pandas_postprocessing, operation)(df, **options)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/superset/utils/pandas_postprocessing/rank.py", line 37, in rank
    df["rank"] = gb.apply(lambda x: x[metric].rank(pct=True))
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/pandas/core/frame.py", line 4081, in __setitem__
    self._set_item_frame_value(key, value)
  File "/home/user/.local/share/pipx/venvs/apache-superset/lib/python3.10/site-packages/pandas/core/frame.py", line 4239, in _set_item_frame_value
    raise ValueError(
ValueError: Cannot set a DataFrame with multiple columns to the single column rank

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]

Thanks for the detailed report! I can confirm this is a bug in the pandas postprocessing rank function.

The issue is in superset/utils/pandas_postprocessing/rank.py at line 37:

if group_by:
    gb = df.groupby(group_by, group_keys=False)
    df["rank"] = gb.apply(lambda x: x[metric].rank(pct=True))

When "Normalize Across X" is selected and a filter reduces the data to a single X value (single column), the groupby on that column produces groups where each group has only one row. In newer versions of pandas, gb.apply(lambda x: x[metric].rank(pct=True)) can return a DataFrame with multiple columns instead of a Series when each group has a single element, causing the ValueError: Cannot set a DataFrame with multiple columns to the single column rank error.

This is a known pandas behavior change — when apply is called on single-element groups, pandas may not properly infer the return type and returns a DataFrame instead of a Series.

A potential fix would be to ensure the result is always a Series before assignment, for example:

if group_by:
    gb = df.groupby(group_by, group_keys=False)
    result = gb.apply(lambda x: x[metric].rank(pct=True))
    if isinstance(result, pd.DataFrame):
        result = result.iloc[:, 0]
    df["rank"] = result
else:
    df["rank"] = df[metric].rank(pct=True)

Or alternatively, using transform instead of apply which always returns a Series aligned with the original DataFrame:

if group_by:
    gb = df.groupby(group_by, group_keys=False)
    df["rank"] = gb[metric].transform(lambda x: x.rank(pct=True))

The transform approach is likely cleaner since it avoids the apply return-type ambiguity altogether and is semantically more appropriate here (we want a value per row, not an aggregation).

For the secondary issue (empty plot when both X and Y are reduced to a single value with normalize across X/Y), that's likely because a single-element group produces a rank of NaN or 1.0 that may not map correctly to the visualization — but fixing the ValueError above should make that case reachable for further debugging.

To reply, just mention @dosu.

---

Share context across your team and agents. Try Dosu.

[EUbaldiEC]

Hi,
The bot pointed me in the (almost) right direction.

Turns out the second line in the if block within rank.py

if group_by:
        gb = df.groupby(group_by, group_keys=False)
        df["rank"] = gb.apply(lambda x: x[metric].rank(pct=True))

will fail when the heatmap has only one row/column.

Substituting it with:

if group_by:
        df["rank"] = df.groupby(group_by)[metric].rank(pct=True)

solves the issue.

Running the old version against the test categories df, but filtered for one category, does indeed raise the error:

> tmp_df = categories_df[categories_df["dept"]=="dept0"].reset_index(drop=True)
> tmp_df.drop(columns=["rank"], errors="ignore", inplace=True)
> rank_upstream(tmp_df, "asc_idx", "dept")  # Prev version
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
....
ValueError: Cannot set a DataFrame with multiple columns to the single column rank

> tmp_df = categories_df[categories_df["dept"]=="dept0"].reset_index(drop=True)
> tmp_df.drop(columns=["rank"], errors="ignore", inplace=True)
> rank(tmp_df, "asc_idx", "dept")  # NEW version
	constant	category	dept	name	asc_idx	desc_idx	idx_nulls	rank
0	dummy	cat0	dept0	person0	0	100	0.0	0.047619
1	dummy	cat2	dept0	person5	5	95	5.0	0.095238
2	dummy	cat1	dept0	person10	10	90	10.0	0.142857
...

I am drafting a PR with all the details and unit tests, but can already show here that this is working as expected in the failing scenarios (while preserving the same behaviour in the ones already functioning).

• Overall case (same works with normalisation across heatmap / y):

• Single column and normalize across X:

• Single row and normalize across Y:

• Single row AND column: