Matplotlib colorbar extensions, show values

Question

I am trying to create a contour so that the user can specify a contour range so that they can they focus on a certain value range. Depending on the data, there will be values outside the user specified range. I adapted this code to illustrate my question:

Python/Matplotlib - Colorbar Range and Display Values

This code plots values from 0 to 2, but I have set the contour range 'v'(0.1 to 1.5)so that there will be values outside the contour and I am using extensions (extend=both) to extend contour so that values outside the range, appear as a colour (to ensure that there are no white areas). What I want is to display the maximum value (2) and minimum value(-0.1) outside the contour range, on the contour scale so that they appear on the scale as well, how can I do that?

On less there is another way

import numpy as np
import matplotlib.pyplot as plt
xi = np.array([0., 0.5, 1.0])
yi = np.array([0., 0.5, 1.0])
zi = np.array([[0., 1.0, 2.0],[0., 1.0, 2.0],[-0.1, 1.0, 2.0]])
v = np.linspace(0.1, 1.5, 10, endpoint=True)#set so that there are values outside contour
plt.contour(xi, yi, zi, v, linewidths=0.5, colors='k')
plt.contourf(xi, yi, zi, v, cmap=plt.cm.jet,extend='both')
cbar=plt.colorbar(ticks=v,extendrect='True')
plt.show()

What I want is for the minimum value of the data and maximum value of the data to display on the extensions.

I have modified the code for the labels of the contour to include the minimum value of the contour, the contour levels and the maximum of the contour. The minimum and maximum values do not appear on the color bar (updated image below) zi.min() is -0.1 and zi.max() is 2.
I investigated the y-ticks: print cbar.ax.get_yticks(): produces values 0 to 1 print cbar.ax.get_ylim(): produces values -0.11 and 1.11 (the ends of the extension bars)

How can I get the minimum and maximum values to appear? can I use the yticks/ylim information?

import numpy as np
import matplotlib.pyplot as plt

xi = np.array([0., 0.5, 1.0])
yi = np.array([0., 0.5, 1.0])
zi = np.array([[0., 1.0, 2.0],
               [0., 1.0, 2.0],
               [-0.1, 1.0, 2.0]])
v = np.linspace(0.1, 1.5, 10, endpoint=True)#contour levels
plt.contour(xi, yi, zi, v, linewidths=0.5, colors='k')
plt.contourf(xi, yi, zi, v, cmap=plt.cm.jet,extend='both')
cbarticks=np.hstack([zi.min(),v,zi.max()])#contour tick labels
cbar=plt.colorbar(ticks=cbarticks,extendrect='True',extendfrac='auto',spacing='proportional')
print cbar.ax.get_ylim()
print cbar.ax.get_yticks()#
plt.show()

Answer 1

Say you have zmin and zmax from the data and umin, umax from user input that wants to focus on a specific range of z values.

You should be able define and use a custom colormap starting from a stdcm=pl.cm.whatever, like this

blue      blue  ----------  red      red
zmin      umin              umax     zmax

and eventually specify explicitly the tics on the colorbar

cbtics = [zmin]+[umin+i*(umax-umin)/n for i in range(n+1)]+[zmax]

Answer 2

xi = np.array([0., 0.5, 1.0])
yi = np.array([0., 0.5, 1.0])
zi = np.array([[0., 1.0, 2.0],[0., 1.0, 2.0],[-0.1, 1.0, 2.0]])
v = np.linspace(0.1, 1.5, 10, endpoint=True)
tick = v.flatten()
tick = np.insert(tick,[0,10],[-0.1, 2.0])
CS = plt.contourf(xi, yi, zi, len(tick), cmap=plt.cm.jet, levels=tick)
C = plt.contour(xi, yi, zi, len(tick), colors='black', levels=tick )
plt.colorbar(CS, shrink=0.5, aspect=5, ticks=tick)
plt.show()