I am working on a GUI where a user can specify the both the min and max x limit. When the value is left blank I would like it to be automatically calculated. One limit can be set while the other is automatically calculated by setting it to None. But after setting one limit and then setting it to None does not automatically update the limit. Both values can be automatically calculated by using set_xlim(auto=True) but this forces both values to be automatically calculated and does not allow only one value to be automatically calculated. Is there a way of having just one limit automatically re-calculated? For example in MATLAB xlim[-Inf, 4] would automatically calculate the first limit.

Below is an example without a GUI

import matplotlib.pyplot as plt

x_data = [1, 2, 3, 4, 5]
y_data = [1, 2, 3, 4, 5]

fig, ax = plt.subplots(1)

ax.plot(x_data, y_data, '.')

# Initially limits are automatically calculated
ax.set_xlim(auto=True)
print(ax.get_xlim()) # prints: (np.float64(0.8), np.float64(5.2))

# User sets value with GUI
ax.set_xlim(2, 4)
print(ax.get_xlim()) # prints: (np.float64(2.0), np.float64(4.0))

# Later user changes their mind and leaves the first limit blank
ax.set_xlim(None, 4) # 
print(ax.get_xlim()) # prints: (np.float64(2.0), np.float64(4.0))

plt.show()

I am working on a GUI where a user can specify the both the min and max x limit. When the value is left blank I would like it to be automatically calculated. One limit can be set while the other is automatically calculated by setting it to None. But after setting one limit and then setting it to None does not automatically update the limit. Both values can be automatically calculated by using set_xlim(auto=True) but this forces both values to be automatically calculated and does not allow only one value to be automatically calculated. Is there a way of having just one limit automatically re-calculated? For example in MATLAB xlim[-Inf, 4] would automatically calculate the first limit.

Below is an example without a GUI

import matplotlib.pyplot as plt

x_data = [1, 2, 3, 4, 5]
y_data = [1, 2, 3, 4, 5]

fig, ax = plt.subplots(1)

ax.plot(x_data, y_data, '.')

# Initially limits are automatically calculated
ax.set_xlim(auto=True)
print(ax.get_xlim()) # prints: (np.float64(0.8), np.float64(5.2))

# User sets value with GUI
ax.set_xlim(2, 4)
print(ax.get_xlim()) # prints: (np.float64(2.0), np.float64(4.0))

# Later user changes their mind and leaves the first limit blank
ax.set_xlim(None, 4) # 
print(ax.get_xlim()) # prints: (np.float64(2.0), np.float64(4.0))

plt.show()

• python
• matplotlib

• stackoverflow/a/66805331/9877065 --> question cited above ; ax.set_ylim(bottom=None, top=7) – pippo1980 Commented Feb 21 at 0:11
• sorry , was to fast voting to reopen , doesnt workk in matplotlib 3.7 too ; once set None doesnt work – pippo1980 Commented Feb 21 at 0:17
• see matplotlib./stable/api/_as_gen/… ----> Passing None leaves the limit unchanged. – pippo1980 Commented Feb 21 at 0:31
• just pass automatically calculated limit to a tuple object --> auto = ax.get_xlim() and then instead of ax.set_xlim(None, 4) use --> ax.set_xlim(auto[0], 4) – pippo1980 Commented Feb 21 at 0:44
• 2 If you call ax.set_xlim(auto=True) again followed by ax.autoscale_view() then it resets to what you originally had. Then you can go ahead and call ax.set_xlim(None, 4)). – RuthC Commented Feb 21 at 11:42

2 Answers 2

As noted in the comment, we can "reset" to automatic axis limits using ax.set_xlim(auto=True) again followed by ax.autoscale_view().

# Later user changes their mind and leaves the first limit blank
ax.set_xlim(auto=True)
ax.autoscale_view()
ax.set_xlim(None, 4)
print(ax.get_xlim()) # prints: (0.8, 4.0)

Got another way to do it, don't now why it works though.

Code:

import matplotlib.pyplot as plt

from matplotlib import __version__

print('\n\nMatplotlib  version  : ', __version__ ,'\n\n')

x_data = [1, 2, 3, 4, 5]
y_data = [1, 2, 3, 4, 5]

fig, ax = plt.subplots(1)

ax.plot(x_data, y_data, '.')

# Initially limits are automatically calculated
ax.set_xlim(auto=True)
print(ax.get_xlim()) # prints: (np.float64(0.8), np.float64(5.2))

# User sets value with GUI
ax.set_xlim(2, 4)
print(ax.get_xlim()) # prints: (np.float64(2.0), np.float64(4.0))

# Later user changes their mind and leaves the first limit blank

print('before None : ', ax.get_xlim())
#ax.set_xlim(auto=True)
#ax.autoscale_view()
#ax.set_xlim(None, 4) #
print('after None : ', ax.get_xlim())
print('before autoscale : ', ax.get_xlim())


ax.autoscale(enable=True, axis='x', tight=None)
ax.set_xlim(None, 4) #
print('after autoscale : ', ax.get_xlim())

print(ax.get_xlim()) # prints: (np.float64(2.0), np.float64(4.0))

plt.show()

Output:

Matplotlib  version  :  3.8.4 


(np.float64(0.8), np.float64(5.2))
(np.float64(2.0), np.float64(4.0))
before None :  (np.float64(2.0), np.float64(4.0))
after None :  (np.float64(2.0), np.float64(4.0))
before autoscale :  (np.float64(2.0), np.float64(4.0))
after autoscale :  (np.float64(0.8), np.float64(4.0))
(np.float64(0.8), np.float64(4.0))