Unexpected result with log-scaled axes

[eltos]

Using views on log-scaled axes results in distorted artists and gives unexpected results compared to plotting the same data twice.

To reproduce, modify the simple example slightly

from matplotview import view
import matplotlib.pyplot as plt

fig, (ax1, ax2) = plt.subplots(1, 2)

# Plot a line in the first axes.
ax1.plot([i for i in range(10)], "-o")

ax1.set(xscale="log", yscale="log")  # <--------- added this line

# Create a view! Turn axes 2 into a view of axes 1.
view(ax2, ax1)
# Modify the second axes data limits so we get a slightly zoomed out view
ax2.set_xlim(-5, 15)
ax2.set_ylim(-5, 15)

fig.show()

Actual result (using view)	Expected result (plotting the data twice)

Passing scale_lines=False to the view improves the situation a bit, but still the markers are distorted:

Actual result (using view with scale_lines=False)	Expected result (plotting the data twice)

Can we have something like scale_artists=False which applies to all artists (markers, text annotations, etc.) and not just lines?
Or even scale=False which applies to everything?

[isaacrobinson2000]

@eltos thanks for reporting this issue! Finally had time to look at this, and it looks like markers were being scaled even when scale_lines=False, which is unintended behavior, I've fixed this in the linked PR. The results above with scale_lines=True is expected, so I'll leave the behavior as-is. Here is the new results with scale_lines=False: