BUG: fix set_zlim method with symlog norm

[neutrinoceros]

PR Summary

fix #3229

import numpy as np
import yt

d = {'my_fake_density': np.random.random(size=(64,64,64)) - 0.5}

ds = yt.load_uniform_grid(d, [64,64,64])
p = yt.SlicePlot(ds, "z", ('stream', 'my_fake_density'))
p.set_log(('stream', 'my_fake_density'), True, linthresh=1.e-3)
p.set_zlim(('stream', 'my_fake_density'), -1e-3, 1e5)
p.save()

on main

This plot is identical before/after calling the set_zlim method.

this branch

PR Checklist

• New features are documented, with docstrings and narrative docs
• Adds a test for any bugs fixed. Adds tests for new features.

[neutrinoceros]

requesting @chummels for review on this one :)

[chummels]

Thank you! I tested this out both with your example as well as my example from Issue #3229 , and they both behave correctly now. I just included some suggestions to remove extra debug print statements.

The only thing that I'll note is that the zmax doesn't appear to show up in as a tickmark/ticklabel in either of the examples. I imagine it could be fixed by slightly bumping the zmax by some epsilon value upward (zmax * 1.01). But other than that this is a great fix.

[neutrinoceros]

The only thing that I'll note is that the zmax doesn't appear to show up in as a tickmark/ticklabel in either of the examples.

that sounds like a reasonable request, however, I anticipate that it'd be very hard to implement this in a way that would work consistently with every matplotlib versions we want to support, as they have been changing the way colorbars work a couple times between 3.0 and 3.5

[chummels]

But won't my option of just slightly increasing the zmax work regardless of the version? It seems like we just want to have the tick label, so making zmax exceed the tick marker value should work for any MPL version.

[chummels]

Maybe bumping down zmin by a similar amount would be wise too? Just to ensure it always has a tick label as well.

[neutrinoceros]

I've seen similar tricks turn into maintenance puzzles with every recent matplotlib release. Matplotlib itself may be the place where this needs fixing. In any case I am very uncomfortable with introducing colorbar tricks in Yt when matplotlib 3.5 is coming up with a complete colorbar overhaul.

[matthewturk]

I'm a little uncertain about removing the nanmax's in favor of straight max's, but I think it looks good.

[neutrinoceros]

I'm a little uncertain about removing the nanmax's in favor of straight max's

I... don't think that's what I'm doing here. I'm actually keeping nanmaxes everywhere, aren't I ?

[matthewturk]

I've marked this for merging once the conflicts are done.

[chummels]

I agree with @neutrinoceros that there are no removed nanmaxes, as it's just defaulting to either nanmax or the value provided by the user (which will not be a nan).

[chummels]

I don't see what conflicts exist or need to be addressed before this gets merged.

[matthewturk]

I don't see what conflicts exist or need to be addressed before this gets merged.

yt/visualization/base_plot_types.py

[neutrinoceros]

fixed !

[neutrinoceros]

@yt-fido test this please

[matthewturk]

@meeseeksdev backport to yt-4.0.x

[neutrinoceros]

@meeseeksdev backport to yt-4.0.x

[neutrinoceros]

Let's try this one last time now that a lot of other PRs went in. If it fails we'll resort to manual backport
@meeseeksdev backport to yt-4.0.x

[lumberbot-app]

Owee, I'm MrMeeseeks, Look at me.

There seem to be a conflict, please backport manually. Here are approximate instructions:

1. Checkout backport branch and update it.

$ git checkout yt-4.0.x
$ git pull

2. Cherry pick the first parent branch of the this PR on top of the older branch:

$ git cherry-pick -m1 71f5772b7f147f21acff614198fd8f12d79390c1

3. You will likely have some merge/cherry-pick conflict here, fix them and commit:

$ git commit -am 'Backport PR #3538: BUG: fix set_zlim method with symlog norm'

4. Push to a named branch :

git push YOURFORK yt-4.0.x:auto-backport-of-pr-3538-on-yt-4.0.x

5. Create a PR against branch yt-4.0.x, I would have named this PR:

"Backport PR #3538 on branch yt-4.0.x (BUG: fix set_zlim method with symlog norm)"

And apply the correct labels and milestones.

Congratulation you did some good work! Hopefully your backport PR will be tested by the continuous integration and merged soon!

Remember to remove Still Needs Manual Backport label once the PR gets merged.

If these instruction are inaccurate, feel free to suggest an improvement.