Bug summary

As of Dec. 11 2023,Axes.pcolorfast() seems to not work withdatetime; see#27450 for more details.

Code for reproduction

importnumpyasnpimportmatplotlib.pyplotaspltimportmatplotlib.datesasdtimportmatplotlibasmplimportdatetimenp.random.seed(19680801)'''Directly inputting datetime.datetime without processing them seems to generate compile errors for pcolorfast().I have used date2num() to ensure that the graph prints correctly despite that such operations may defy the purpose of conducting such tests.'''basedate_x=datetime.datetime(2023,12,6,1,30,30)basedate_y=datetime.datetime(2024,5,5,12,15,45)dates_x= [(basedate_x+datetime.timedelta(days=1*i,hours=6*i,minutes=20*i,seconds=0))foriinrange(10)]dates_y= [(basedate_y+datetime.timedelta(days=1*i,hours=8*i,minutes=30*i,seconds=0))foriinrange(10)]data=np.random.rand(0,100)fig,ax=plt.subplots()pc=ax.pcolorfast(dates_x,dates_y,data)ax.set_xlabel('Sample datetime')ax.set_ylabel('Sample data')ax.set_title('Sample test case for pcolorfast()')plt.show()

Actual outcome

Traceback (most recent call last):
File "personal directory omitted", line 23, in
pc = ax.pcolorfast(dates_x, dates_y, data)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\andyx\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\LocalCache\local-packages\Python311\site-packages\matplotlib_init_.py", line 1478, in inner
return func(ax, *map(sanitize_sequence, args), **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\andyx\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\LocalCache\local-packages\Python311\site-packages\matplotlib\axes_axes.py", line 6501, in pcolorfast
self.update_datalim(np.array([[xl, yb], [xr, yt]]))
File "C:\Users\andyx\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\LocalCache\local-packages\Python311\site-packages\matplotlib\axes_base.py", line 2510, in update_datalim
if not np.any(np.isfinite(xys)):
^^^^^^^^^^^^^^^^
TypeError: ufunc 'isfinite' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''

Expected outcome

It has been assumed that a graph should be produced similar to the one below, but with actualdatetime as axes:

Additional information

This is part of a test case for issue#26864 ; same issue has been met when the same code is ran on matplotlib ver 3.7.1 on Google Colab.

Operating system

Windows 10

Matplotlib Version

3.8.2

Matplotlib Backend

No response

Python version

3.11.7

Jupyter version

No response

Installation

pip