[ENH] : Provide axis('equal') for Axes3D #22570 by patquem · Pull Request #22705 · matplotlib/matplotlib

patquem · 2022-03-25T10:35:44Z

PR Summary

Closes #22570
Make axis('equal') available
(I am new in PR - sorry for not doing things correctly)

PR Checklist

Tests and Styling

Has pytest style unit tests (and pytest passes).
Is Flake 8 compliant (install flake8-docstrings and run flake8 --docstring-convention=all).

Documentation

New features are documented, with examples if plot related.
New features have an entry in doc/users/next_whats_new/ (follow instructions in README.rst there).
[N/A] API changes documented in doc/api/next_api_changes/ (follow instructions in README.rst there).
Documentation is sphinx and numpydoc compliant (the docs should build without error).

Example

import numpy as np
import matplotlib.pyplot as plt


def plot_hexahedron(ax, x, y, z, *args, **kwargs):
    def plot_edge(i, j):
        ax.plot([x[i], x[j]], [y[i], y[j]], [z[i], z[j]], *args, **kwargs)

    # edges associated to the y=0 face
    plot_edge(0, 1)
    plot_edge(1, 3)
    plot_edge(3, 2)
    plot_edge(2, 0)

    # edges associated to the y=1 face
    plot_edge(4, 5)
    plot_edge(5, 7)
    plot_edge(7, 6)
    plot_edge(6, 4)

    # missing edges (x=cste)
    plot_edge(0, 4)
    plot_edge(1, 5)
    plot_edge(3, 7)
    plot_edge(2, 6)


def ex_hexahedron():
    """ Example of hexahedron representation with Matplotlib """
    points = np.array([[0, 0, 0], [1, 0, 0], [0, 1, 0], [1, 1, 0],
                       [0, 0, 3], [1, 0, 3], [0, 1, 3], [1, 1, 3]])

    x = np.asarray([coord[0] for coord in points])
    y = np.asarray([coord[1] for coord in points])
    z = np.asarray([coord[2] for coord in points])

    fig, axes = plt.subplots(1, 2, figsize=(12, 5),
                             subplot_kw={'projection': '3d'})

    def plot(ax):
        plot_hexahedron(ax, x, y, z, 'r')
        ax.set_xlabel("X")
        ax.set_ylabel("Y")
        ax.set_zlabel("Z")

    # standard representation (aspect='auto')
    plot(axes[0])
    axes[0].set_title("Standard representation")

    # 3d axis equal (aspect='equal')
    plot(axes[1])
    axes[1].set_aspect('equal')
    axes[1].set_title("set_aspect('equal)")
    plt.tight_layout()


ex_hexahedron()
plt.show()

github-actions

Thank you for opening your first PR into Matplotlib!

If you have not heard from us in a while, please feel free to ping @matplotlib/developers or anyone who has commented on the PR. Most of our reviewers are volunteers and sometimes things fall through the cracks.

You can also join us on gitter for real-time discussion.

For details on testing, writing docs, and our review process, please see the developer guide

We strive to be a welcoming and open project. Please follow our Code of Conduct.

oscargus · 2022-03-25T15:01:24Z

Thanks!

The failure comes from a test that tests that equal is not supported...

matplotlib/lib/mpl_toolkits/tests/test_mplot3d.py

Lines 24 to 28 in 5971feb

    
           def test_aspect_equal_error(): 
        
               fig = plt.figure() 
        
               ax = fig.add_subplot(projection='3d') 
        
               with pytest.raises(NotImplementedError): 
        
                   ax.set_aspect('equal')

So this should be replaced with some test testing the outcome of setting aspect='equal'. For example, something similar to the one that you have.

For the actual implementation, I cannot really say much as I do not know the part of this code very well. One can note that it partly seems to work, but that it may also need some adjustments. For example, the 2.0 value of the x-axis is cropped and it looks like the hexahedron is from 0.5 to 1.5 in y-direction.

patquem · 2022-03-29T09:29:58Z

Hi @oscargus,
I have changed the test.
I agree with you concerning the visualization (confusing in the Y-direction) but I have no idea of how to fix that.
Patrick

oscargus · 2022-04-01T13:59:59Z

It seems like it is the location of the ytick-labels that is off. The -1.0 label is outside of the viewable area and all other are shifted one position. So I guess that the actual box is drawn correctly, but the labels are misplaced.

Edit: To clarify, I guess that this problem is not caused by your change, so it seems to be an issue elsewhere.

patquem · 2022-04-01T14:14:43Z

I agree with you.

Concerning the test, I will try to finalize it next week.
Presently, the figure I gave for comparison was generated with my current working env.
This engenders some differences with figure generated by the main branch.
I have to generate a new figure from my forked (and updated) matplotlib branch, to be coherent with it, I guess.

oscargus · 2022-04-05T10:34:59Z

            value       description
            =========   ==================================================
            'auto'      automatic; fill the position rectangle with data.
+            'equal'     automatic; adapt the axes to have equal aspect


Suggested change

'equal' automatic; adapt the axes to have equal aspect

'equal' automatic; adapt the axes to have equal aspect.

oscargus · 2022-04-05T10:35:35Z

        mpl_toolkits.mplot3d.axes3d.Axes3D.set_box_aspect
        """
-        if aspect != 'auto':
+        if aspect not in ['auto', 'equal']:


Suggested change

if aspect not in ['auto', 'equal']:

if aspect not in ('auto', 'equal'):

Minor thing, but slightly faster this way as this can be "compiled".

oscargus · 2022-04-05T10:40:27Z

Approved subject to minor fixes and rebasing/squashing into single commit.

oscargus · 2022-04-05T10:41:09Z

The error comes from adding a figure in one commit and updating in another. If you rebase and squash the commits into a single one, it should go away.

oscargus · 2022-04-05T10:44:55Z

Oh, you are using the main branch. In general it is better to not do that, but to create a separate branch.

Now, you need to do something like
git rebase -i HEAD~4
(HEAD minus four commits)
change pick to fixup for the last three.
git push --force-with-lease

Then, for later updates you can do

git commit --amend

so that the change will stay in that single commit.

oscargus · 2022-04-05T14:43:52Z

We can also squash the commits when merging.

patquem · 2022-04-05T16:49:20Z

Hi @oscargus ,
It sounds like chinese for me.
I have tried to do what you wrote. Not sure I did the things correctly. (?)

oscargus · 2022-04-07T07:02:50Z

I think that may not have worked, but do not worry, we can sort it out while merging.

github-actions Bot reviewed Mar 25, 2022

View reviewed changes

oscargus added the topic: mplot3d label Mar 25, 2022

oscargus reviewed Apr 5, 2022

View reviewed changes

oscargus approved these changes Apr 5, 2022

View reviewed changes

oscargus added the status: needs workflow approval For PRs from new contributors, from which GitHub blocks workflows by default. label Apr 5, 2022

patquem mentioned this pull request May 9, 2022

[ENH] : Provide axis('equal') for Axes3D (replace PR #22705) #23017

Closed

6 tasks

patquem closed this May 9, 2022

patquem force-pushed the main branch from e5b492e to 6b2d3f2 Compare May 9, 2022 10:06

QuLogic added the status: duplicate label May 9, 2022

liuzhenqi77 mentioned this pull request Nov 17, 2022

Update aspect parameter in plot_point_brain netneurolab/netneurotools#122

Open

	'equal' automatic; adapt the axes to have equal aspect
	'equal' automatic; adapt the axes to have equal aspect.

	if aspect not in ['auto', 'equal']:
	if aspect not in ('auto', 'equal'):

Uh oh!

Conversation

patquem commented Mar 25, 2022 • edited by oscargus Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Summary

PR Checklist

Example

Uh oh!

github-actions Bot left a comment

Choose a reason for hiding this comment

Uh oh!

oscargus commented Mar 25, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

patquem commented Mar 29, 2022

Uh oh!

oscargus commented Apr 1, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

patquem commented Apr 1, 2022

Uh oh!

oscargus Apr 5, 2022

Choose a reason for hiding this comment

Uh oh!

oscargus Apr 5, 2022

Choose a reason for hiding this comment

Uh oh!

oscargus commented Apr 5, 2022

Uh oh!

oscargus commented Apr 5, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

oscargus commented Apr 5, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

oscargus commented Apr 5, 2022

Uh oh!

patquem commented Apr 5, 2022

Uh oh!

oscargus commented Apr 7, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

patquem commented Mar 25, 2022 •

edited by oscargus

Loading

oscargus commented Mar 25, 2022 •

edited

Loading

oscargus commented Apr 1, 2022 •

edited

Loading

oscargus commented Apr 5, 2022 •

edited

Loading

oscargus commented Apr 5, 2022 •

edited

Loading