networkx中具有“关键”节点可访问的最短路径

Question

I have a directed graph in python G developed with networkx. The graph has weights called 'weight'.

I know an explicit start node A and an End Node F. In between the graph can access the nodes B,C,D,E.

How can i explicity say that he has to acces B and D by finding the shortest path and can addionally add C and E, if this helps to the shortest path?

So far I know the function:

nx.single_source_dijkstra(G, 'A', target='F', cutoff=None, weight='weight')

which gives the output:

(10.01,
['A',
 'B',
 'C',
 'F',])

How can I make sure that it includes E?

Answer 1

Networkx has no built-in functions or arguments for your problem. You should do it manually:

import networkx as nx

# Create a random DAG
G = nx.gnp_random_graph(50,0.3,directed=True)
DAG = nx.DiGraph([(u,v) for (u,v) in G.edges() if u<v])
nx.is_directed_acyclic_graph(DAG)
for edge in G.edges:
    G.edges[edge]['weight'] = 1

# Get the longest path (without weights) from node 1 to node 40
# with nodes 5, 10, 20, 30 inside
max([
    (path, len(path))
    for path in nx.all_simple_paths(DAG, 1, 40)
    if all(n in path for n in (5, 10, 20, 30))
], key=lambda x: x[1])

# Get the longest path (with weights)
max([
    path
    for path in nx.all_simple_paths(DAG, 1, 40)
    if all(n in path for n in (5, 10, 20, 30))
], key=lambda x: sum(G.edges[edge]['weight'] for edge in nx.utils.pairwise(x)))