Fix(plotting): Bug fixes for issues #587 and #588

autotest/t007_test.py

@@ -406,11 +406,11 @@ def test_mt_modelgrid():
     mt = flopy.mt3d.Mt3dms(modelname='test_mt', modflowmodel=ml,
                            model_ws=ml.model_ws, verbose=True)
     btn = flopy.mt3d.Mt3dBtn(mt, icbund=ml.bas6.ibound.array)
-    
+
     # reload swt
     swt = flopy.seawat.Seawat(modelname='test_swt', modflowmodel=ml,
                               mt3dmodel=mt, model_ws=ml.model_ws, verbose=True)
-    
+
     assert \
         ml.modelgrid.xoffset == mt.modelgrid.xoffset == swt.modelgrid.xoffset
     assert \
@@ -903,6 +903,42 @@ def check_vertices():
     plt.close()
 
 
+def test_tricontour_NaN():
+    from flopy.plot import PlotMapView
+    import numpy as np
+    from flopy.discretization import StructuredGrid
+
+    arr = np.random.rand(10, 10) * 100
+    arr[-1, :] = np.nan
+    delc = np.array([10] * 10, dtype=float)
+    delr = np.array([8] * 10, dtype=float)
+    top = np.ones((10, 10), dtype=float)
+    botm = np.ones((3, 10, 10), dtype=float)
+    botm[0] = 0.75
+    botm[1] = 0.5
+    botm[2] = 0.25
+    idomain = np.ones((3, 10, 10))
+    idomain[0, 0, :] = 0
+    vmin = np.nanmin(arr)
+    vmax = np.nanmax(arr)
+    levels = np.linspace(vmin, vmax, 7)
+
+    grid = StructuredGrid(delc=delc,
+                          delr=delr,
+                          top=top,
+                          botm=botm,
+                          idomain=idomain,
+                          lenuni=1,
+                          nlay=3, nrow=10, ncol=10)
+
+    pmv = PlotMapView(modelgrid=grid, layer=0)
+    contours = pmv.contour_array(a=arr)
+
+    for ix, lev in enumerate(contours.levels):
+        if not np.allclose(lev, levels[ix]):
+            raise AssertionError("TriContour NaN catch Failed")
+
+
 def test_get_vertices():
     from flopy.utils.reference import SpatialReference
     from flopy.discretization import StructuredGrid
@@ -1191,7 +1227,7 @@ def test_export_array_contours():
     # build_sfr_netcdf()
     # test_mg()
     # test_mbase_modelgrid()
-    test_mt_modelgrid()
+    # test_mt_modelgrid()
     # test_rotation()
     # test_model_dot_plot()
     # test_vertex_model_dot_plot()
@@ -1212,5 +1248,6 @@ def test_export_array_contours():
     #test_wkt_parse()
     #test_get_rc_from_node_coordinates()
     # test_export_array()
-    # test_export_array_contours()
+    test_export_array_contours()
+    test_tricontour_NaN()
     pass

flopy/plot/map.py

@@ -125,7 +125,13 @@ def plot_array(self, a, masked_values=None, **kwargs):
                 raise Exception('Array must be of dimension 1, 2, or 3')
 
         elif self.mg.grid_type == "vertex":
-            if a.ndim == 2:
+            if a.ndim == 3:
+                if a.shape[0] == 1:
+                    a = np.squeeze(a, axis=1)
+                    plotarray = a[self.layer, :]
+                else:
+                    raise Exception("Array must be of dimension 1 or 2")
+            elif a.ndim == 2:
                 plotarray = a[self.layer, :]
             elif a.ndim == 1:
                 plotarray = a
@@ -215,6 +221,7 @@ def contour_array(self, a, masked_values=None, **kwargs):
             err_msg = "Matplotlib must be updated to use contour_array"
             raise ImportError(err_msg)
 
+        a = np.copy(a)
         if not isinstance(a, np.ndarray):
             a = np.array(a)
 
@@ -232,7 +239,13 @@ def contour_array(self, a, masked_values=None, **kwargs):
                 raise Exception('Array must be of dimension 1, 2 or 3')
 
         elif self.mg.grid_type == "vertex":
-            if a.ndim == 2:
+            if a.ndim == 3:
+                if a.shape[0] == 1:
+                    a = np.squeeze(a, axis=1)
+                    plotarray = a[self.layer, :]
+                else:
+                    raise Exception("Array must be of dimension 1 or 2")
+            elif a.ndim == 2:
                 plotarray = a[self.layer, :]
             elif a.ndim == 1:
                 plotarray = a
@@ -242,9 +255,39 @@ def contour_array(self, a, masked_values=None, **kwargs):
         else:
             plotarray = a
 
+        # work around for tri-contour ignore vmin & vmax
+        # necessary block for tri-contour NaN issue
+        if "levels" not in kwargs:
+            if "vmin" not in kwargs:
+                vmin = np.nanmin(plotarray)
+            else:
+                vmin = kwargs.pop("vmin")
+            if "vmax" not in kwargs:
+                vmax = np.nanmax(plotarray)
+            else:
+                vmax = kwargs.pop('vmax')
+
+            levels = np.linspace(vmin, vmax, 7)
+            kwargs['levels'] = levels
+
+        # workaround for tri-contour nan issue
+        # use -2**31 to allow for 32 bit int arrays
+        plotarray[np.isnan(plotarray)] = -2**31
+        if masked_values is None:
+            masked_values = [-2**31]
+        else:
+            masked_values = list(masked_values)
+            if -2**31 not in masked_values:
+                masked_values.append(-2**31)
+
+        ismasked = None
         if masked_values is not None:
             for mval in masked_values:
-                plotarray = np.ma.masked_equal(plotarray, mval)
+                if ismasked is None:
+                    ismasked = np.equal(plotarray, mval)
+                else:
+                    t = np.equal(plotarray, mval)
+                    ismasked += t
 
         if 'ax' in kwargs:
             ax = kwargs.pop('ax')
@@ -276,16 +319,11 @@ def contour_array(self, a, masked_values=None, **kwargs):
         ycentergrid = ycentergrid.flatten()
         triang = tri.Triangulation(xcentergrid, ycentergrid)
 
-        mask = None
-        try:
-            amask = plotarray.mask
-            mask = [False for i in range(triang.triangles.shape[0])]
-            for ipos, (n0, n1, n2) in enumerate(triang.triangles):
-                if amask[n0] or amask[n1] or amask[n2]:
-                    mask[ipos] = True
+        if ismasked is not None:
+            ismasked = ismasked.flatten()
+            mask = np.any(np.where(ismasked[triang.triangles],
+                                   True, False), axis=1)
             triang.set_mask(mask)
-        except (AttributeError, IndexError):
-            pass
 
         contour_set = ax.tricontour(triang, plotarray, **kwargs)
 

flopy/plot/plotutil.py

@@ -1833,6 +1833,56 @@ def line_intersect_grid(ptsin, xgrid, ygrid):
 
         return vdict
 
+    @staticmethod
+    def irregular_shape_patch(xverts, yverts):
+        """
+        Patch for vertex cross section plotting when
+        we have an irregular shape type throughout the
+        model grid or multiple shape types.
+
+        Parameters
+        ----------
+        xverts : list
+            xvertices
+        yverts : list
+            yvertices
+
+        Returns
+        -------
+            xverts, yverts as np.ndarray
+
+        """
+        max_verts = 0
+
+        for xv in xverts:
+            if len(xv) > max_verts:
+                max_verts = len(xv)
+
+        for yv in yverts:
+            if len(yv) > max_verts:
+                max_verts = len(yv)
+
+        adj_xverts = []
+        for xv in xverts:
+            if len(xv) < max_verts:
+                n = max_verts - len(xv)
+                adj_xverts.append(xv + [xv[-1]] * n)
+            else:
+                adj_xverts.append(xv)
+
+        adj_yverts = []
+        for yv in yverts:
+            if len(yv) < max_verts:
+                n = max_verts - len(yv)
+                adj_yverts.append(yv + [yv[-1]] * n)
+            else:
+                adj_yverts.append(yv)
+
+        xverts = np.array(adj_xverts)
+        yverts = np.array(adj_yverts)
+
+        return xverts, yverts
+
     @staticmethod
     def arctan2(verts):
         """

flopy/plot/vcrosssection.py

@@ -90,11 +90,22 @@ def __init__(self, ax=None, model=None, modelgrid=None,
                                self.mg.xoffset, self.mg.yoffset,
                                self.mg.angrot_radians, inverse=True)
 
-        self.xvertices, self.yvertices = \
-            geometry.transform(self.mg.xvertices,
-                               self.mg.yvertices,
-                               self.mg.xoffset, self.mg.yoffset,
-                               self.mg.angrot_radians, inverse=True)
+        try:
+            self.xvertices, self.yvertices = \
+                geometry.transform(self.mg.xvertices,
+                                   self.mg.yvertices,
+                                   self.mg.xoffset, self.mg.yoffset,
+                                   self.mg.angrot_radians, inverse=True)
+        except ValueError:
+            # irregular shapes in vertex grid ie. squares and triangles
+            xverts, yverts = plotutil.UnstructuredPlotUtilities.\
+                irregular_shape_patch(self.mg.xvertices, self.mg.yvertices)
+
+            self.xvertices, self.yvertices = \
+                geometry.transform(xverts, yverts,
+                                   self.mg.xoffset,
+                                   self.mg.yoffset,
+                                   self.mg.angrot_radians, inverse=True)
 
         pts = [(xt, yt) for xt, yt in zip(xp, yp)]
         self.pts = np.array(pts)
@@ -430,9 +441,38 @@ def contour_array(self, a, masked_values=None, head=None, **kwargs):
         plotarray = np.array([a[cell] for cell
                               in sorted(self.projpts)])
 
+        # work around for tri-contour ignore vmin & vmax
+        # necessary for the tri-contour NaN issue fix
+        if "levels" not in kwargs:
+            if "vmin" not in kwargs:
+                vmin = np.nanmin(plotarray)
+            else:
+                vmin = kwargs.pop("vmin")
+            if "vmax" not in kwargs:
+                vmax = np.nanmax(plotarray)
+            else:
+                vmax = kwargs.pop('vmax')
+
+            levels = np.linspace(vmin, vmax, 7)
+            kwargs['levels'] = levels
+
+        # workaround for tri-contour nan issue
+        plotarray[np.isnan(plotarray)] = -2**31
+        if masked_values is None:
+            masked_values = [-2**31]
+        else:
+            masked_values = list(masked_values)
+            if -2**31 not in masked_values:
+                masked_values.append(-2**31)
+
+        ismasked = None
         if masked_values is not None:
             for mval in masked_values:
-                plotarray = np.ma.masked_equal(plotarray, mval)
+                if ismasked is None:
+                    ismasked = np.equal(plotarray, mval)
+                else:
+                    t = np.equal(plotarray, mval)
+                    ismasked += t
 
         if isinstance(head, np.ndarray):
             zcenters = self.set_zcentergrid(np.ravel(head))
@@ -457,15 +497,11 @@ def contour_array(self, a, masked_values=None, head=None, **kwargs):
 
         triang = tri.Triangulation(xcenters, zcenters)
 
-        try:
-            amask = plotarray.mask
-            mask = [False for _ in range(triang.triangles.shape[0])]
-            for ipos, (n0, n1, n2) in enumerate(triang.triangles):
-                if amask[n0] or amask[n1] or amask[n2]:
-                    mask[ipos] = True
+        if ismasked is not None:
+            ismasked = ismasked.flatten()
+            mask = np.any(np.where(ismasked[triang.triangles],
+                                   True, False), axis=1)
             triang.set_mask(mask)
-        except (AttributeError, IndexError):
-            pass
 
         contour_set = ax.tricontour(triang, plotarray, **kwargs)
 

flopy/utils/geometry.py

@@ -360,9 +360,9 @@ def transform(x, y, xoff, yoff, angrot_radians,
 
     """
     if isinstance(x, list):
-        x = np.array(x)
+        x = np.array(x, dtype=float)
     if isinstance(y, list):
-        y = np.array(y)
+        y = np.array(y, dtype=float)
 
     if not np.isscalar(x):
         x, y = x.copy(), y.copy()