b = self.boundaries[self._inside]

locator = ticker.FixedLocator(b)

if self.values is None and self.boundaries is not None:

# define the outside additions

if self.extend == 'min':

to_add = self._boundaries[0]

elif self.extend == 'max':

to_add = self._boundaries[-1]

elif self.extend == 'both':

to_add = self._boundaries[[0, -1]]

elif self.extend == 'neither':

to_add = []

# add values to the colorbar insides

boundaries_inside = self._boundaries[self._inside]

n_between = max((self.cmap.N-len(to_add)) //

(len(boundaries_inside)-1), 1)

self._values = np.linspace(boundaries_inside[:-1],

boundaries_inside[1:],

n_between, endpoint=False,

axis=-1).flatten()

self._values = np.append(self._values, boundaries_inside[-1])

# add back in the boundaries for the extensions

self._boundaries = np.sort(np.concatenate([to_add,

self._values]))

# calculate values at which to plot colors

self._values = 0.5 * (self._values[:-1] + self._values[1:])

# add back in the boundaries for the extensions, display

# colors of extensions at boundary values

self._values = np.sort(np.concatenate([to_add,

self._values]))

if isinstance(self.norm, colors.NoNorm):

self._values = (self._values + 0.00001).astype(np.int16)

elif self.values is None and self.boundaries is None: