Bug summary

In #20752 I added support for imshow(..., norm="log"), and documented that the feature is supported in particular by hexbin as well; but this doesn't actually work for hexbin, because hexbin tries to run stuff on the norm before normalizing strings.

Code for reproduction

from pylab import *
hexbin(rand(100), rand(100), norm="log")

Actual outcome

  File ".../matplotlib/axes/_axes.py", line 5247, in hexbin
    if norm.vmin is None and norm.vmax is None:
       ^^^^^^^^^
AttributeError: 'str' object has no attribute 'vmin'

Expected outcome

no error

Additional information

Probably the block

        # autoscale the norm with current accum values if it hasn't been set
        if norm is not None:
            if norm.vmin is None and norm.vmax is None:
                norm.autoscale(accum)

can just be moved down a bit after collection.set_norm(norm) (then we can get back the canonicalized norm from the collection). I haven't looked at how this interacts with the call to _scale_norm just below, though.

Operating system

any

Matplotlib Version

3.9.0.dev1523+g2723052176

Matplotlib Backend

any

Python version

3.12

Jupyter version

no

Installation

None