sharex or sharey with aspect=equal affects axis limits

[godaygo]

When aspect='equal' and sharex/sharey are used together, they affect the axis limits, and I can not precisely set them.

fig = plt.figure()

ax1 = fig.add_subplot(1,2,1)
ax2 = fig.add_subplot(1,2,2, sharey=ax1)

ax1.plot([.1, .5, .9], [.1, .5, .9], 'o')
ax2.plot([.1, .5, .9], [.1, .5, .9], 'o')

ax1.axis([0, 1, 0, 1])
ax2.axis([0, 1, 0, 1])

ax1.set_xticks([0, .2, .4, .6, .8, 1])
ax2.set_xticks([0, .2, .4, .6, .8, 1])

plt.show()

Also sharex and sharey behaves differently. Above picture was for sharey, below is for sharex:

When I use one of them everything looks fine, with aspect=equal, without share:

fig = plt.figure()

ax1 = fig.add_subplot(1,2,1, aspect='equal')
ax2 = fig.add_subplot(1,2,2, aspect='equal')

ax1.plot([.1, .5, .9], [.1, .5, .9], 'o')
ax2.plot([.1, .5, .9], [.1, .5, .9], 'o')

ax1.axis([0, 1, 0, 1])
ax2.axis([0, 1, 0, 1])

ax1.set_xticks([0, .2, .4, .6, .8, 1])
ax2.set_xticks([0, .2, .4, .6, .8, 1])

plt.show()

without aspect=equal, with sharey:

Also the trouble is the same for set_aspect and axis('equal').

matplotlib 1.5.3.
python 3.5.2

[tacaswell]

This is because you are trying to over-constrain the system. You essentially have the equations

x_range = C1
y_range = C2
x_range =  λ * y_range

which clearly can not be solved if C1 = λ*C2 .

[godaygo]

@tacaswell Sorry, I can not fully understand your idea. 'equal' means that the length of line segment will be the same on x and y, and share only shares. How they both here affect axis limits?

About over-determined systems, as I remember they can have zero, one, or infinity number of solutions. Here as I see, the solution is: x_range=[0, 1]; y_range=[0, 1]; λ =1. But as I say I don't understand how does this situation relate with over-determined systems? I have YAxis of ax2 to share with YAxis of ax1. The aspect must be equal (the corresponding lengths on axes) and nothing more. But the Xaxis limits become broken.

I think that the picture should be (like with no sharing):

[godaygo]

@tacaswell Accoording to this issue #7654. If I use adjustable='box-forced' kwarg, everything works fine as expected.

fig = plt.figure()
ax1 = fig.add_subplot(121, adjustable='box-forced', aspect='equal')
ax2 = fig.add_subplot(122, adjustable='box-forced', aspect='equal', sharey=ax1)

ax1.plot([.1, .5, .9], [.1, .5, .9], 'o')
ax2.plot([.1, .5, .9], [.1, .5, .9], 'o')

ax1.axis([0, 1, 0, 1])
ax2.axis([0, 1, 0, 1])

ax1.set_xticks([0, .2, .4, .6, .8, 1])
ax2.set_xticks([0, .2, .4, .6, .8, 1])

plt.show()

[github-actions]

This issue has been marked "inactive" because it has been 365 days since the last comment. If this issue is still present in recent Matplotlib releases, or the feature request is still wanted, please leave a comment and this label will be removed. If there are no updates in another 30 days, this issue will be automatically closed, but you are free to re-open or create a new issue if needed. We value issue reports, and this procedure is meant to help us resurface and prioritize issues that have not been addressed yet, not make them disappear. Thanks for your help!