def test_all_out_edges_of_node(self):
graph = DiGraph()
for i in range(10):
graph.add_node(i)
for i in range(9):
graph.add_edge(i, i + 1, i * 2 + 1)
self.assertTrue(graph.all_out_edges_of_node(0).__contains__(1))
graph.remove_edge(0, 1)
self.assertFalse(graph.all_in_edges_of_node(0).__contains__(1))
graph.add_node(12)
graph.add_edge(12, 4, 19)
self.assertIn(4, graph.all_out_edges_of_node(12))
def test_remove_edge(self):
graph = DiGraph()
for i in range(10):
graph.add_node(i)
for i in range(9):
graph.add_edge(i, i + 1, i * 2 + 1)
graph.remove_edge(2, 3)
self.assertEqual(graph.v_size(), 10)
self.assertEqual(graph.e_size(), 8)
self.assertFalse(graph.all_out_edges_of_node(2).__contains__(3))
self.assertFalse(graph.all_in_edges_of_node(3).__contains__(2))
self.assertFalse(graph.remove_edge(6, 0))
self.assertFalse(graph.remove_edge(5, 100))
def check0():
"""
This function tests the naming (main methods of the DiGraph class, as defined in GraphInterface.
:return:
"""
g = DiGraph() # creates an empty directed graph
for n in range(4):
g.add_node(n)
g.add_edge(0, 1, 1)
g.add_edge(1, 0, 1.1)
g.add_edge(1, 2, 1.3)
g.add_edge(2, 3, 1.1)
g.add_edge(1, 3, 1.9)
g.remove_edge(1, 3)
g.add_edge(1, 3, 10)
print(g.v_size()) # prints the __repr__ (func output)
print(g.e_size()) # prints the __repr__ (func output)
print(g.get_all_v()) # prints a dict with all the graph's vertices.
print(g.all_in_edges_of_node(1))
print(g.all_out_edges_of_node(1))
g_algo = GraphAlgo(g)
print(g_algo.shortest_path(0, 3))
g_algo.plot_graph()
class GraphAlgo(GraphAlgoInterface):
def __init__(self, graph: DiGraph = DiGraph()):
self.graph = graph
self.string = "Graph"
def get_graph(self) -> GraphInterface:
return self.graph
def load_from_json(self, file_name: str) -> bool:
self.string = file_name
vertex = {}
edges = {}
try:
with open(file_name, "r") as file:
new_graph_dict = json.load(file)
vertex = new_graph_dict["Nodes"]
edges = new_graph_dict["Edges"]
self.graph = DiGraph()
for i in vertex:
if "pos" not in i:
i["pos"] = None
self.graph.add_node(i["id"], i["pos"])
for e in edges:
self.graph.add_edge(e["src"], e["dest"], e["w"])
except IOError as e:
print(e)
return False
return True
def save_to_json(self, file_name: str) -> bool:
vertex = []
edges = []
graph = {}
for i in self.graph.get_all_v():
n = {"pos": i.pos, "id": i.key}
vertex.append(n)
for j in self.graph.all_out_edges_of_node(i.key).values():
e = {"src": i.key, "w": j.weight, "dest": j.dst}
edges.append(e)
graph["Edges"] = edges
graph["Nodes"] = vertex
try:
with open(file_name, "w") as file:
json.dump(graph, indent=4, fp=file)
except IOError as e:
print(e)
return False
return True
def shortest_path(self, id1: int, id2: int) -> (float, list):
if id1 not in self.graph.Nodes.keys(
) or id2 not in self.graph.Nodes.keys():
return float('infinity'), []
self.Dijkstra(id1)
ls = []
if self.graph.Nodes.get(id2).tag == float('infinity'):
return float('infinity'), ls
form = id2
ls.append(form)
while id1 not in ls:
ls.append(self.graph.Nodes.get(form).r_from.key)
form = self.graph.Nodes.get(form).r_from.key
ls.reverse()
return self.graph.Nodes.get(id2).tag, ls
def connected_component(self, id1: int):
if id1 not in self.graph.Nodes.keys() or self.graph is None:
return []
l_ist = self.Kosaraju(id1)
return l_ist
def connected_components(self) -> List[list]:
if self.graph is None:
return []
l_ist = []
al_inScc = []
for i in self.graph.Nodes.values():
if i.key not in al_inScc:
l = self.Kosaraju(i.key)
for k in l:
al_inScc.append(k)
l_ist.append(l)
return l_ist
def plot_graph(self) -> None:
plt.title(self.string)
x_vals = []
y_vals = []
for xy in self.graph.Nodes.values():
if xy.pos is None:
xy.pos = (random.randrange(0, 100), random.randrange(0,
100), 0)
x_vals.append(xy.pos[0])
y_vals.append(xy.pos[1])
else:
string = xy.pos.split(',')
xy.pos = tuple(string)
x_vals.append(float(string[0]))
y_vals.append(float(string[1]))
plt.plot(x_vals, y_vals, 'o')
for v in self.graph.Nodes.values():
v_x = float(v.pos[0])
v_y = float(v.pos[1])
plt.annotate(v.key, (v_x - 0.00015, v_y + 0.00015), color='red')
for e in self.graph.EdgesSrc[v.key].values():
x_e = float(self.graph.Nodes[e.dst].pos[0])
y_e = float(self.graph.Nodes[e.dst].pos[1])
plt.arrow(v_x,
v_y,
x_e - v_x,
y_e - v_y,
length_includes_head=True,
head_width=0.0001991599,
width=0.0000005,
color='blue')
plt.show()
def Dijkstra(self, src):
"""
Dijkstra Algorithm :
Traverse the Graph from Source Node and tag each reachable Node with the MinSum Weight
Use's for Shortest_Path
"""
for vertex in self.graph.Nodes.values():
vertex.tag = float('infinity')
self.graph.Nodes.get(src).tag = 0
pq = [self.graph.Nodes.get(src)]
visited = [src]
while len(pq) > 0:
node = pq.pop(0)
for neighbor in self.graph.EdgesSrc[node.key].values():
weight = node.tag + neighbor.weight
if weight < self.graph.Nodes.get(neighbor.dst).tag:
self.graph.Nodes.get(neighbor.dst).tag = weight
self.graph.Nodes.get(neighbor.dst).r_from = node
if neighbor.dst not in visited:
pq.append(self.graph.Nodes.get(neighbor.dst))
visited.append(neighbor.dst)
return
def Kosaraju(self, src):
"""
This Algorithm is for finding all SCC in the graph -> Use on all Vertices
Work this way ->
DFS on the Graph - Contain a list of all reachable Vertices from Source Node
Graph^T = Transpose Original Graph
DFS on the Graph^T - Contain a list of all reachable Vertices from Source Node (On Graph^T)
:return An Intersection between both lists - > The Source Node SCC
"""
s = [src]
visited = {}
while len(s) > 0: # DFS Graph
v = s.pop()
if v not in visited.keys():
visited[v] = self.graph.Nodes[v]
for edge in self.graph.EdgesSrc[v].values():
s.append(edge.dst)
visited_2 = {}
s_2 = [src]
while len(s_2) > 0:
v = s_2.pop()
if v not in visited_2.keys():
visited_2[v] = self.graph.Nodes[v]
for edge in self.graph.EdgesDst[v].values():
s_2.append(edge.dst)
x = set(visited).intersection(visited_2)
return list(x)