clabel randomly inconsistend when placed manually

[dcrabs]

Setting more than 4 labels manually with clabel in contourplot results in "random" distribution of labels across contours.

import numpy as np
import matplotlib.pyplot as plt

a = 0.2
resolution = 100
xarray = np.linspace(0,0.25,num=resolution)
yarray = np.linspace(0,1,num=resolution)

A = np.empty([resolution,resolution])

xc = 0
yc = 0
for x in xarray:
    for y in yarray:
        #print xc,yc

        wp = 1./np.pi*np.arctan(np.sin(np.pi*y)*np.sinh(np.pi*a/2.)/
              (np.cosh(np.pi*x)-np.cos(np.pi*y)*np.cosh(np.pi*a/2.)))

        if wp <= 0:
            wp = wp+1
            A[xc, yc] = wp
        else:
            A[xc, yc] = wp
        yc += 1
    yc=0
    xc += 1

A = A.transpose()
B = np.fliplr(A)
AB = np.hstack((B,A))
fullx = np.hstack((-xarray[::-1],xarray))

#plot
CS = plt.contour(fullx,yarray,AB,10, colors='k')

labelpos = np.dstack((np.zeros(9),np.arange(0.1,1,0.1)))[0]
plt.clabel(CS,inline=True, fmt='%1.1f',fontsize=9, manual=labelpos)

plt.show()

[dcrabs]

Well, it seems it is the y axis that is relevant not the value of the contour.

[tacaswell]

As a bit of clairification:

manual:

if True, contour labels will be placed manually using mouse clicks. Click the first button near a contour to add a label, click the second button (or potentially both mouse buttons at once) to finish adding labels. The third button can be used to remove the last label added, but only if labels are not inline. Alternatively, the keyboard can be used to select label locations (enter to end label placement, delete or backspace act like the third mouse button, and any other key will select a label location).

manual can be an iterable object of x,y tuples. Contour labels will be created as if mouse is clicked at each x,y positions.

So it is putting labels as close as possible to the x,y locations as possible (and sorting out which contour and it's label) internally.