PR Summary

The original code for reproduction of this issue #23817 is

import matplotlib.pyplot as plt

fig, ax = plt.subplots()
cs = ax.contour([[1, 1], [1, 1]])
fig.colorbar(cs, ax=ax)
plt.show()

However, this seems to yield the following TypeError
TypeError: Input z must be 2D, not 1D

The same TypeError also occurs if we change

cs = ax.contour([[1, 1], [1, 1]])

to

cs = ax.contour([[1, 2], [3, 4]])

This TypeError is resulted from lib/matplotlib/contour.py, line 1526, in _initialize_x_y, which makes me think that rather than a colorbar problem, the root of this issue lies in how contour works. What is the expected behavior of the original code for reproduction? If we aren't supposed to call Fontour on input like above in the first place, how is Colorbar supposed to behave and why does the the API warning even matter?

I was able to suppress the failed test case with the following change to @quinnah's test case from #24656

def test_contour_uniformfield_colorbar():
    # Smoke test for issue
    fig, ax = plt.subplots()
    with pytest.warns(Warning) as record:
        cs = ax.contour([[2, 2], [2, 2]])
    assert len(record) == 1
    try:
        fig.colorbar(cs, ax=ax)
    except:
        pass

PR Checklist

Documentation and Tests

• Has pytest style unit tests (and pytest passes)
• Documentation is sphinx and numpydoc compliant (the docs should build without error).
• New plotting related features are documented with examples.

Release Notes

• New features are marked with a .. versionadded:: directive in the docstring and documented in doc/users/next_whats_new/
• API changes are marked with a .. versionchanged:: directive in the docstring and documented in doc/api/next_api_changes/
• Release notes conform with instructions in next_whats_new/README.rst or next_api_changes/README.rst