Graph.from_networkx method now extract node and edge attributes (#2714)

.travis.yml

@@ -27,7 +27,7 @@ install:
   - conda update -q conda
   # Useful for debugging any issues with conda
   - conda info -a
-  - conda create -q -n test-environment python=$TRAVIS_PYTHON_VERSION flake8 scipy=1.0.0 numpy freetype nose pandas=0.22.0 jupyter ipython=5.4.1 param matplotlib=2.1.2 xarray
+  - conda create -q -n test-environment python=$TRAVIS_PYTHON_VERSION flake8 scipy=1.0.0 numpy freetype nose pandas=0.22.0 jupyter ipython=5.4.1 param matplotlib=2.1.2 xarray networkx
   - source activate test-environment
   - conda install -c conda-forge  iris plotly flexx ffmpeg netcdf4=1.3.1 --quiet
   - conda install -c bokeh datashader dask bokeh=0.12.15 selenium

holoviews/element/graphs.py

@@ -385,10 +385,17 @@ def from_networkx(cls, G, layout_function, nodes=None, **kwargs):
         """
         Generate a HoloViews Graph from a networkx.Graph object and
         networkx layout function. Any keyword arguments will be passed
-        to the layout function.
+        to the layout function. By default it will extract all node
+        and edge attributes from the networkx.Graph but explicit node
+        information may also be supplied.
         """
         positions = layout_function(G, **kwargs)
-        edges = G.edges()
+        edges = []
+        for start, end in G.edges():
+            attrs = sorted(G.adj[start][end].items())
+            edges.append((start, end)+tuple(v for k, v in attrs))
+        edge_vdims = [k for k, v in attrs] if edges else []
+
         if nodes:
             idx_dim = nodes.kdims[-1].name
             xs, ys = zip(*[v for k, v in sorted(positions.items())])
@@ -398,8 +405,13 @@ def from_networkx(cls, G, layout_function, nodes=None, **kwargs):
             nodes = nodes.add_dimension('x', 0, xs)
             nodes = nodes.add_dimension('y', 1, ys).clone(new_type=cls.node_type)
         else:
-            nodes = cls.node_type([tuple(pos)+(idx,) for idx, pos in sorted(positions.items())])
-        return cls((edges, nodes))
+            nodes = []
+            for idx, pos in sorted(positions.items()):
+                attrs = sorted(G.nodes[idx].items())
+                nodes.append(tuple(pos)+(idx,)+tuple(v for k, v in attrs))
+            vdims = [k for k, v in attrs] if nodes else []
+            nodes = cls.node_type(nodes, vdims=vdims)
+        return cls((edges, nodes), vdims=edge_vdims)
 
 
 

tests/element/testgraphelement.py

@@ -2,6 +2,7 @@
 Unit tests of Graph Element.
 """
 from unittest import SkipTest
+from nose.plugins.attrib import attr
 
 import numpy as np
 from holoviews.core.data import Dataset
@@ -128,8 +129,45 @@ def test_graph_redim_nodes(self):
         self.assertEqual(redimmed.nodes, graph.nodes.redim(x='x2', y='y2'))
         self.assertEqual(redimmed.edgepaths, graph.edgepaths.redim(x='x2', y='y2'))
 
-
-
+    @attr(optional=1)
+    def test_from_networkx_with_node_attrs(self):
+        try:
+            import networkx as nx
+        except:
+            raise SkipTest('Test requires networkx to be installed')
+        G = nx.karate_club_graph()
+        graph = Graph.from_networkx(G, nx.circular_layout)
+        clubs = np.array([
+            'Mr. Hi', 'Mr. Hi', 'Mr. Hi', 'Mr. Hi', 'Mr. Hi', 'Mr. Hi',
+            'Mr. Hi', 'Mr. Hi', 'Mr. Hi', 'Officer', 'Mr. Hi', 'Mr. Hi',
+            'Mr. Hi', 'Mr. Hi', 'Officer', 'Officer', 'Mr. Hi', 'Mr. Hi',
+            'Officer', 'Mr. Hi', 'Officer', 'Mr. Hi', 'Officer', 'Officer',
+            'Officer', 'Officer', 'Officer', 'Officer', 'Officer', 'Officer',
+            'Officer', 'Officer', 'Officer', 'Officer'])
+        self.assertEqual(graph.nodes.dimension_values('club'), clubs)
+
+    @attr(optional=1)
+    def test_from_networkx_with_edge_attrs(self):
+        try:
+            import networkx as nx
+        except:
+            raise SkipTest('Test requires networkx to be installed')
+        FG = nx.Graph()
+        FG.add_weighted_edges_from([(1,2,0.125), (1,3,0.75), (2,4,1.2), (3,4,0.375)])
+        graph = Graph.from_networkx(FG, nx.circular_layout)
+        self.assertEqual(graph.dimension_values('weight'), np.array([0.125, 0.75, 1.2, 0.375]))
+
+    @attr(optional=1)
+    def test_from_networkx_only_nodes(self):
+        try:
+            import networkx as nx
+        except:
+            raise SkipTest('Test requires networkx to be installed')
+        G = nx.Graph()
+        G.add_nodes_from([1, 2, 3])
+        graph = Graph.from_networkx(G, nx.circular_layout)
+        self.assertEqual(graph.nodes.dimension_values(2), np.array([1, 2, 3]))
+
 class ChordTests(ComparisonTestCase):
 
     def setUp(self):