I like the general approach more here.

Yes, I agree, this is much better! :)

In 515aa53, I remove the if artist.stale:, because when using blitting, the animated artists which are not stale will be missing, because they will either: not be part of the background or not be drawn (because stale is False!).

Adapting one of the blitting example:

import matplotlib.pyplot as plt
from matplotlib.widgets import SpanSelector
import numpy as np

x = np.linspace(0, 2 * np.pi, 100)
values = np.sin(x)

fig, ax = plt.subplots()
(ln,) = ax.plot(x, values, animated=True)
(ln2 , ) = ax.plot([], animated=True)

plt.show(block=False)
plt.pause(0.1)

bg = fig.canvas.copy_from_bbox(fig.bbox)
ax.draw_artist(ln)
fig.canvas.blit(fig.bbox)

def mean(vmin, vmax):
    indmin, indmax = np.searchsorted(x, (vmin, vmax))
    v = values[indmin:indmax].mean()
    ln2.set_data(x, v)
    
span = SpanSelector(ax, mean, direction='horizontal',
                    onmove_callback=mean,
                    interactive=True,
                    drag_from_anywhere=True,
                    useblit=True)

I have added a test and switch to list comprehension instead of using filter.

2a5c86c fixes an issue with zorder of the artists. Before 2a5c86c, the selector artists will be drawn before other artists and therefore it was behind, as in the following example:

import matplotlib.pyplot as plt
from matplotlib.widgets import SpanSelector
import numpy as np

values = np.ones((10, 10))

fig, ax = plt.subplots()
ax.imshow(values, animated=True)

span = SpanSelector(ax, print, direction='horizontal',
                    interactive=True,
                    drag_from_anywhere=True,
                    useblit=True)

when sorting the artists order, all artists need to sorted together!

@dstansby, any chance you can have a look at this again, please? Thanks!

This PR is affected by a re-writing of our history to remove a large number of accidentally committed files see discourse for details.

To recover this PR it will need be rebased onto the new default branch (main). There are several ways to accomplish this, but we recommend (assuming that you call the matplotlib/matplotlib remote "upstream"

git remote update
git checkout main
git merge --ff-only upstream/main
git checkout YOUR_BRANCH
git rebase --onto=main upstream/old_master
# git rebase -i main # if you prefer
git push --force-with-lease   # assuming you are tracking your branch

If you do not feel comfortable doing this or need any help please reach out to any of the Matplotlib developers. We can either help you with the process or do it for you.

Thank you for your contributions to Matplotlib and sorry for the inconvenience.

Can anyone review this PR, please?

@dstansby, if this PR looks good to you - from your comment in #21342 (review); can you please approve and merge? :) Thanks!

Rebased to get the test suite to pass, can anyone merge this PR?