Tim Hoffmann

Thanks @erykoff and congratulations on your first contribution to Matplotlib.

I think the casing c vs C is not consistent in our docs. In general, we tend to use uppercase for 2D arrays (think `X, Y = np.meshgrid(x, x)`), e.g....

@meeseeksdev backport to v3.6.x This is not v3.6-doc because it contains changes to files in`lib` (that it‘s only docstrings there does not matter).

@meeseeksdev backport to v3.6.x

Unfortunately, selectors are not supported in `environment.yml` https://github.com/conda/conda/issues/8089

I think pip (with a comment) is viable. If that's too messy, simply add the condo package but commented out, plus a hint that Linux/win users could activate it.

I'm also surprised, but this > I turned out, however, that practically in all cases, where the output is intentional, it was already produced by print (or by error messages)...

This seems sort of intentional: https://github.com/matplotlib/matplotlib/blob/2c5c3518e8d23412b6ca2009ddc26f2edacd5600/lib/matplotlib/axis.py#L1189 @anntzer can you comment on that? This seems like a regression from #10088. Anyway, if this is the desired behavior, we should add a...

Should we add an include_hidden parameter to control the behavior?

Can you please rebase to confirm that all (likely unrelated) CI problems are gone?