Antony Lee
Antony Lee
My personal preference would be for radians, because that matches what numpy uses basically everywhere (even in things such as np.unwrap, which are not bound by consistency with C functions...
Not a very big deal, but this does fix a bug relatively easily so I'll nudge for a review to get this included in 3.8.
Sure, your point is well taken; the issue is that right now, this "map to the centerpoint" only happens if you add a colorbar to the image: ```python from pylab...
Deprecating singular norms seems OK too, as long as we don't warn on `imshow(single-value-array)` (i.e. `autoscale(single-value-array)` should just silently autoexpand): not warning is consistent with `plot(single-value)` also silently autoexpanding (which...
Couldn't that just be done by adapting FileMovieWriter accordingly? In one of my oldest PRs (#3536) I argued that FileMovieWriter should default to putting its files in a tempdir anyways;...
It seems natural enough to convert everything to the axis' tz?
@timhoffm I had started giving a try to the idea (not pursuing it actively now), per https://github.com/anntzer/matplotlib/tree/_wip/newaxis (taking an approach of "delete everything and reimplement stuff on top of it")....
#5804 is possibly relevant re: formatter work too.
One case (not that I use it, but doesn't sounds too crazy) is https://matplotlib.org/examples/api/skewt.html, where you can see that for some of the slanted gridlines, the associated tick is not...