Pandas enforces subplot_adjust creating UserWarning with constrained_layout

[michaelaye]

Bug report

Bug summary

Pandas dataframe plotting uses subplot_adjust, always creating a UserWarning because I am using constrained_layout as my default layouter.

Cross-issue with pandas-dev/pandas#25261 (I'm uncertain where this should/could be fixed).

Code for reproduction

import matplotlib as mpl
mpl.rcParams['figure.constrained_layout.use'] = True
tseries = pd.date_range('now', freq='s', periods=100)
df = pd.DataFrame({'data': np.random.random(100)}, index=tseries)
df.plot()

Actual outcome

Expected outcome

No UserWarning

Matplotlib version

• Operating system: macOS 10.14.6
• Matplotlib version: 3.1.3
• Matplotlib backend (print(matplotlib.get_backend())): 'module://ipykernel.pylab.backend_inline'
• Python version: 3.7.6
• Jupyter version (if applicable): Jupyterlab 1.2.6
• Other libraries:

All installed via conda-forge.

[jklymak]

This is how constrained_layout is supposed to work - if the user manually adjusts the subplots, it stops trying to do things automatically.

[michaelaye]

With user you mean pandas here, I guess?

[jklymak]

Any user or any package calling subplots_adjust.

[michaelaye]

Ok. My point here is that it's out of the user's hand. Pandas just does the subplots_adjust, therefore any user of the constrained_layout feature constantly gets these UserWarnings.

[jklymak]

Matplotlib is working as designed, so I guess the question is what change would you hope for on our end? The only two I can think of are a) we don't emit a UserWarning, but I think the warning is needed in this case, or b) we ignore subplots_adjust. But I think we can't really do that as people have perfectly valid reasons for manually adjusting their plots, and don't want some more general setting overriding the more specific.

From the panda's side they could not call subplots_adjust if constrained_layout is True.

[michaelaye]

I tend to agree, but the matplotlib twitter account encouraged me to create this issue and link it to the pandas' one. ;)

[story645]

I figured this is a common enough bug that it's useful to be documented here 🤷‍♀

[github-actions]

This issue has been marked "inactive" because it has been 365 days since the last comment. If this issue is still present in recent Matplotlib releases, or the feature request is still wanted, please leave a comment and this label will be removed. If there are no updates in another 30 days, this issue will be automatically closed, but you are free to re-open or create a new issue if needed. We value issue reports, and this procedure is meant to help us resurface and prioritize issues that have not been addressed yet, not make them disappear. Thanks for your help!

[story645]

This doesn't seem to happen anymore due to changes in Matplotlib or Pandas.