Additionally checked for the case when the desired vmax or vmin are outside the range of the data. In this case, the contour plot, if extend for the appropriate direction is also chosen, will appropriately reflect the desired range.

kthyng · kthyng · commit d806c1a803f6 · 2013-06-28T17:23:12.000-05:00
diff --git a/lib/matplotlib/contour.py b/lib/matplotlib/contour.py
@@ -21,6 +21,7 @@
 import matplotlib.patches as mpatches
 import matplotlib.texmanager as texmanager
 import matplotlib.transforms as mtrans
+import pdb
 
 # Import needed for adding manual selection capability to clabel
 from matplotlib.blocking_input import BlockingContourLabeler
@@ -1196,7 +1197,7 @@ def _process_levels(self):
         # want to leave the original levels attribute unchanged.
         # (Colorbar needs this even for line contours.)
         self._levels = list(self.levels)
-
+        pdb.set_trace()
         if self.extend in ('both', 'min'):
             self._levels.insert(0, min(self.levels[0], self.zmin) - 1)
         if self.extend in ('both', 'max'):
@@ -1495,18 +1496,24 @@ def _contour_args(self, args, kwargs):
                             (fn, fn))
 
         # Check for vmin and vmax values and max out z values accordingly
+        # want zmax and zmin to be able to be outside data range, according to input vmin/vmax
         if 'vmax' in kwargs:
-            vmax = kwargs.pop('vmax')
+            vmax = kwargs['vmax']
             ind = z > vmax
             z[ind] = vmax
+            self.zmax = kwargs.pop('vmax')   
+        else:  
+            self.zmax = ma.maximum(z)
         if 'vmin' in kwargs:
-            vmin = kwargs.pop('vmin')
+            vmin = kwargs['vmin']
             ind = z < vmin
             z[ind] = vmin
+            self.zmin = kwargs.pop('vmin')   
+        else:
+            self.zmin = ma.minimum(z)
 
         z = ma.masked_invalid(z, copy=False)
-        self.zmax = ma.maximum(z)
-        self.zmin = ma.minimum(z)
+
         if self.logscale and self.zmin <= 0:
             z = ma.masked_where(z <= 0, z)
             warnings.warn('Log scale: values of z <= 0 have been masked')
