Parameterize close-node edge creation on new nodes (#30)

* parameterize graph methods

* thread through param

* len check

* drop prints

* update tests

* udpate nodes df before it goes into cross feed checker

* nameify sid

* update tests

* drop commented out code

peartree/graph.py

@@ -56,6 +56,7 @@ def generate_summary_graph_elements(feed: ptg.gtfs.feed,
 
 
 def generate_cross_feed_edges(G: nx.MultiDiGraph,
+                              name: str,
                               stops_df: pd.DataFrame,
                               exempt_nodes: List[str],
                               connection_threshold: float) -> pd.DataFrame:
@@ -69,7 +70,8 @@ def generate_cross_feed_edges(G: nx.MultiDiGraph,
     node_df = generate_graph_node_dataframe(G)
 
     # Remove all nodes that are part of the new additions to the graph
-    node_df = node_df[~node_df.index.isin(exempt_nodes)]
+    if len(exempt_nodes) > 0:
+        node_df = node_df[~node_df.index.isin(exempt_nodes)]
 
     stop_ids = []
     to_nodes = []
@@ -79,6 +81,7 @@ def generate_cross_feed_edges(G: nx.MultiDiGraph,
     #       be a way to condense these two steps well
     for i, row in stops_df.iterrows():
         sid = str(row.stop_id)
+        full_sid = nameify_stop_id(name, sid)
 
         # Ensure that each value is typed correctly prior to being
         # fed into the nearest node method
@@ -88,7 +91,7 @@ def generate_cross_feed_edges(G: nx.MultiDiGraph,
         nearest_nodes = get_nearest_nodes(node_df,
                                           point,
                                           connection_threshold,
-                                          exempt_id=sid)
+                                          exempt_id=full_sid)
 
         # Iterate through series results and add to output
         for node_id, dist_val in nearest_nodes.iteritems():
@@ -171,7 +174,8 @@ def populate_graph(G: nx.MultiDiGraph,
                    wait_times_by_stop: pd.DataFrame,
                    summary_edge_costs: pd.DataFrame,
                    connection_threshold: Union[int, float],
-                   walk_speed_kmph: float=4.5):
+                   walk_speed_kmph: float=4.5,
+                   exempt_internal_edge_imputation: bool=False):
     # Generate a merge of the wait time data and the feed stops data that will
     # be used for both the addition of new stop nodes and the addition of
     # cross feed edges later on (that join one feeds stops to the other if
@@ -182,8 +186,10 @@ def populate_graph(G: nx.MultiDiGraph,
     sid_lookup = _add_nodes_and_edges(G, name, stops_df, summary_edge_costs)
 
     # Generate cross feed edge values
-    exempt_nodes = sid_lookup.values()
-    cross_feed_edges = generate_cross_feed_edges(G, stops_df,
+    exempt_nodes = []
+    if exempt_internal_edge_imputation:
+        exempt_nodes = sid_lookup.values()
+    cross_feed_edges = generate_cross_feed_edges(G, name, stops_df,
                                                  exempt_nodes,
                                                  connection_threshold)
 
@@ -198,7 +204,8 @@ def make_synthetic_system_network(
         name: str,
         reference_geojson: Dict,
         connection_threshold: Union[int, float],
-        walk_speed_kmph: float=4.5):
+        walk_speed_kmph: float=4.5,
+        exempt_internal_edge_imputation: bool=False):
     # Same as populate_graph, we use this dict to monitor the stop ids
     # that are created
     sid_lookup = {}
@@ -237,8 +244,10 @@ def make_synthetic_system_network(
             all_nodes = all_nodes.append(nodes)
 
     # Generate cross feed edge values
-    exempt_nodes = sid_lookup.values()
-    cross_feed_edges = generate_cross_feed_edges(G, all_nodes,
+    exempt_nodes = []
+    if exempt_internal_edge_imputation:
+        exempt_nodes = sid_lookup.values()
+    cross_feed_edges = generate_cross_feed_edges(G, name, all_nodes,
                                                  exempt_nodes,
                                                  connection_threshold)
     # Mutates the G network object

peartree/paths.py

@@ -56,7 +56,8 @@ def load_feed_as_graph(feed: ptg.gtfs.feed,
                        existing_graph: nx.MultiDiGraph=None,
                        connection_threshold: float=50.0,
                        walk_speed_kmph: float=4.5,
-                       interpolate_times: bool=True):
+                       interpolate_times: bool=True,
+                       exempt_internal_edge_imputation: bool=True):
     """
     Convert a feed object into a NetworkX Graph, connect to an existing
     NetworkX graph if one is supplied
@@ -131,15 +132,17 @@ def load_feed_as_graph(feed: ptg.gtfs.feed,
                           wait_times_by_stop,
                           summary_edge_costs,
                           connection_threshold,
-                          walk_speed_kmph)
+                          walk_speed_kmph,
+                          exempt_internal_edge_imputation)
 
 
 def load_synthetic_network_as_graph(
         reference_geojson: Dict,
         name: str=None,
         existing_graph: nx.MultiDiGraph=None,
         connection_threshold: float=50.0,
-        walk_speed_kmph: float=4.5):
+        walk_speed_kmph: float=4.5,
+        exempt_internal_edge_imputation: bool=False):
 
     # Generate a random name for name if it is None
     if not name:
@@ -164,4 +167,5 @@ def load_synthetic_network_as_graph(
         name,
         reference_geojson,
         connection_threshold,
-        walk_speed_kmph)
+        walk_speed_kmph,
+        exempt_internal_edge_imputation)