[j-brendel]

Tests crash with AttributeError: 'Series' object has no attribute 'view'

Problem

Running pytest tests can fail with:

AttributeError: 'Series' object has no attribute 'view'

The traceback points into tests/test_consumption.py::TestConsumption::test_calculate_consumption:

consumption_df = pd.read_csv(self.consumption_path)
consumption_col = consumption_df["consumption_kwh_per_km"].view()
...
consumption_col[:] = true_cons(lol_col, incline_col, speed_col, temp_col)

pandas.Series.view() was deprecated in pandas 2.x and has been removed entirely in pandas 3.x
(installed here: pandas 3.0.5). The call was used to obtain a view of the DataFrame column that
could be mutated in place (consumption_col[:] = ...), so the new values end up back in
consumption_df without reassigning the column explicitly. consumption_col itself is never
used again after that line — the DataFrame is what gets passed on to
consumption_calculator.set_consumption_interpolation(...).

This is purely a test-environment issue (pandas version used in the local venv), not a bug in
simba/consumption.py itself.

Solution

Skip the view/in-place-mutation detour and assign the computed values straight to the column:

consumption_df = pd.read_csv(self.consumption_path)
lol_col = consumption_df["level_of_loading"]
incline_col = consumption_df["incline"]
speed_col = consumption_df["mean_speed_kmh"]
temp_col = consumption_df["t_amb"]

consumption_df["consumption_kwh_per_km"] = true_cons(
    lol_col, incline_col, speed_col, temp_col)

• Functionally identical: consumption_df["consumption_kwh_per_km"] ends up holding the same
  values before being passed into set_consumption_interpolation.
• Works regardless of pandas version, since it no longer depends on the removed .view() API.
• No production code is affected; the fix is contained to the test file.

Implementation (branch fix/pandas_view)

• tests/test_consumption.py: replaced .view() + in-place assignment with a direct column
  assignment in test_calculate_consumption.

[stefansc1]

Also works in pandas 2.x