def test_add_vertex(self):
g = Graph()
g.add_vertex('A')
g.add_vertex('B')
g.add_vertex('C')
g.add_vertex('D')
assert g.has_vertex('A') is True
assert g.has_vertex('B') is True
assert g.has_vertex('C') is True
assert g.has_vertex('D') is True
assert g.has_vertex('E') is False
assert g.has_vertex('F') is False
assert g.has_vertex('G') is False
assert g.has_vertex('H') is False
g.add_edge('A', 'B')
g.add_edge('C', 'D')
self.assertCountEqual(g.get_neighbors('A'), ['B'])
self.assertCountEqual(g.get_neighbors('B'), [])
self.assertCountEqual(g.get_neighbors('C'), ['D'])
self.assertCountEqual(g.get_neighbors('D'), [])
g.add_edge('A', 'F')
g.add_edge('B', 'E')
self.assertCountEqual(g.get_neighbors('A'), ['B', 'F'])
self.assertCountEqual(g.get_neighbors('B'), ['E'])
def _generate_graph(self):
"""
This function generates a graph representing the airline network.
:return :The graph representing the network.
"""
with open(self.routes_file, encoding="utf8") as csv_file:
csv_reader = csv.reader(csv_file, delimiter=',')
g = Graph()
for line in csv_reader:
#decode srcfrom file
src_airport_cord = line[3]
dst_airport_cord = line[5]
if src_airport_cord in self.cordinates and dst_airport_cord in self.cordinates:
d = self._haversine(self.cordinates[src_airport_cord],
self.cordinates[dst_airport_cord])
#add_edge(from_node, to_node, weight, stops, airline_code)
g.add_edge(line[2], line[4], d, line[7], line[0])
#decode src from file
# print(self.cordinates.get(self.source.lower()))
# print(self.source)
if self.cordinates.get(self.source.lower(),
None) == line[3]:
self.source_code = line[2]
self.source_airline = line[0]
#decode dest from file
if self.cordinates.get(self.destination.lower(),
None) == line[5]:
self.destination_code = line[4]
return g
def read_graph_from_file(file_name):
'''Read graph from a file'''
file = open(file_name, 'r')
file_list = file.readlines()
vertices = file_list[1].rstrip().split(',')
valid_types = 'gGdD'
graph_type = ''
directed = False
weighted = False
edges_list = []
if file_list[0].rstrip() in valid_types:
graph_type = file_list[0].upper()
else:
raise ValueError('G or D is not specified')
if graph_type == 'D':
directed = True
for item in range(2, len(file_list)):
edge = file_list[item].rstrip().replace('(', '').replace(')', '').split(',')
if len(edge) == 3:
weighted = True
edges_list.append(edge)
graph = Graph(weighted, directed)
for edge in edges_list:
if weighted:
graph.add_weighted_edge(int(edge[0]), int(edge[1]), int(edge[2]))
else:
graph.add_edge(int(edge[0]), int(edge[1]))
return graph
def test_has_vertex(self):
g = Graph()
assert g.has_vertex('A') is False
g.add_vertex('B')
assert g.has_vertex('B') is True
g.add_edge('C', 'D')
assert g.has_vertex('C') is True
assert g.has_vertex('D') is True
def test_get_neighbors(self):
g = Graph()
g.add_edge('A', 'B')
g.add_edge('C', 'D')
g.add_edge('E', 'F')
g.add_edge('G', 'H')
self.assertCountEqual(g.get_neighbors('A'), ['B'])
self.assertCountEqual(g.get_neighbors('B'), [])
self.assertCountEqual(g.get_neighbors('C'), ['D'])
self.assertCountEqual(g.get_neighbors('D'), [])
with self.assertRaises(KeyError):
g.get_neighbors('Z')
def build_graph(directed):
g = Graph(directed)
vertices = []
for val in ['a', 'b', 'c', 'd', 'e', 'f', 'g']:
vertex = Vertex(val)
vertices.append(vertex)
g.add_vertex(vertex)
for v in range(len(vertices)):
v_idx = randrange(0, len(vertices) - 1)
v1 = vertices[v_idx]
v_idx = randrange(0, len(vertices) - 1)
v2 = vertices[v_idx]
g.add_edge(v1, v2, randrange(1, 10))
print_graph(g)
def test_size(self):
g = Graph()
assert g.size == 0
g.add_vertex('A')
g.add_vertex('B')
g.add_vertex('C')
g.add_vertex('D')
assert g.size == 4
g.add_edge('A', 'B')
g.add_edge('C', 'D')
g.add_edge('E', 'F')
g.add_edge('G', 'H')
assert g.size == 8
with self.assertRaises(KeyError):
g.add_vertex('A')
assert g.size == 8
x = [graph.vertices[vertex_id].x for vertex_id in graph.vertices]
y = [graph.vertices[vertex_id].y for vertex_id in graph.vertices]
graph_layout = dict(zip(node_indices, zip(x, y)))
graph_renderer.layout_provider = StaticLayoutProvider(graph_layout=graph_layout)
plot.renderers.append(graph_renderer)
labelSource = ColumnDataSource(data=dict(x=x, y=y, names=[graph.vertices[vertex_id].value for vertex_id in graph.vertices]))
labels = LabelSet(x='x', y='y', text='names', level='glyph',
text_align='center', text_baseline='middle', source=labelSource, render_mode='canvas')
plot.add_layout(labels)
output_file('graph.html')
show(plot)
graph = Graph() # Instantiate your graph
graph.add_vertex('0')
graph.add_vertex('1')
graph.add_vertex('2')
graph.add_vertex('3')
graph.add_edge('0', '1')
graph.add_edge('0', '3')
graph.add_edge('1', '2')
smth = BokehGraph(graph)
smth.draw()
[3, 3, 2, 1, 1]
]
non_fullfilling = [[4, 3, 2, 2, 2, 1, 1],
[6, 6, 5, 4, 4, 2, 1],
[3, 3, 3, 1]]
for sequence in fullfilling + non_fullfilling:
print(sequence, Graph.erdoes_gallai(sequence))
print("Add vertex 'z':")
graph.add_vertex("z")
print(graph)
print("Add edge ('x','y'): ")
graph.add_edge(('x', 'y'))
print(graph)
print("Add edge ('a','d'): ")
graph.add_edge(('a', 'd'))
print(graph)
ct[start_vertex]:
if vertex not in vertices_encountered:
if self.is_connected(vertices_encountered, vertex):
return True
else:
return True
return False
def vertex_degree(self, vertex):
""" The degree of a vertex is the number of edges connecting
class Test(unittest.TestCase):
def setUp(self):
self.graph = Graph()
self.graph.add_vertex(1)
self.graph.add_vertex(2)
self.graph.add_vertex(3)
self.graph.add_vertex(4)
self.graph.add_vertex(5)
self.graph.add_vertex(6)
self.graph.add_vertex(7)
self.graph.add_edge(5, 3)
self.graph.add_edge(6, 3)
self.graph.add_edge(7, 1)
self.graph.add_edge(4, 7)
self.graph.add_edge(1, 2)
self.graph.add_edge(7, 6)
self.graph.add_edge(2, 4)
self.graph.add_edge(3, 5)
self.graph.add_edge(2, 3)
self.graph.add_edge(4, 6)
def test_vertices(self):
vertices = {
1: {2},
2: {3, 4},
3: {5},
4: {6, 7},
5: {3},
6: {3},
7: {1, 6}
}
self.assertDictEqual(self.graph.vertices, vertices)
def test_bft(self):
bft = [
"1\n2\n3\n4\n5\n6\n7\n", "1\n2\n3\n4\n5\n7\n6\n",
"1\n2\n3\n4\n6\n7\n5\n", "1\n2\n3\n4\n6\n5\n7\n",
"1\n2\n3\n4\n7\n6\n5\n", "1\n2\n3\n4\n7\n5\n6\n",
"1\n2\n4\n3\n5\n6\n7\n", "1\n2\n4\n3\n5\n7\n6\n",
"1\n2\n4\n3\n6\n7\n5\n", "1\n2\n4\n3\n6\n5\n7\n",
"1\n2\n4\n3\n7\n6\n5\n", "1\n2\n4\n3\n7\n5\n6\n"
]
stdout_ = sys.stdout
sys.stdout = io.StringIO()
self.graph.bft(1)
output = sys.stdout.getvalue()
self.assertIn(output, bft)
sys.stdout = stdout_ # Restore stdout
def test_dft(self):
dft = [
"1\n2\n3\n5\n4\n6\n7\n", "1\n2\n3\n5\n4\n7\n6\n",
"1\n2\n4\n7\n6\n3\n5\n", "1\n2\n4\n6\n3\n5\n7\n"
]
stdout_ = sys.stdout
sys.stdout = io.StringIO()
self.graph.dft(1)
output = sys.stdout.getvalue()
self.assertIn(output, dft)
sys.stdout = stdout_ # Restore stdout
def test_dft_recursive(self):
dft = [
"1\n2\n3\n5\n4\n6\n7\n", "1\n2\n3\n5\n4\n7\n6\n",
"1\n2\n4\n7\n6\n3\n5\n", "1\n2\n4\n6\n3\n5\n7\n"
]
stdout_ = sys.stdout
sys.stdout = io.StringIO()
self.graph.dft_recursive(1)
output = sys.stdout.getvalue()
self.assertIn(output, dft)
sys.stdout = stdout_ # Restore stdout
def test_bfs(self):
bfs = [1, 2, 4, 6]
self.assertListEqual(self.graph.bfs(1, 6), bfs)
def test_dfs(self):
dfs = [[1, 2, 4, 6], [1, 2, 4, 7, 6]]
self.assertIn(self.graph.dfs(1, 6), dfs)
def test_dfs_recursive(self):
dfs = [[1, 2, 4, 6], [1, 2, 4, 7, 6]]
self.assertIn(self.graph.dfs_recursive(1, 6), dfs)
def main():
try:
with open(sys.argv[1], 'r') as tweets, open(sys.argv[2], 'w') as output:
# make a pretty global dictionary to store time and hashtag info
hashtag_dict = OrderedDict()
result = '0.00'
# iterate over the input file, one line at a time
for line in tweets:
tweet = line.decode('utf-8')
tweet_json = json.loads(tweet) # dict object
if 'created_at' in tweet_json: # needed to address keyError
# get the time of the tweet
created_at = datetime.strptime(
tweet_json['created_at'].encode('utf-8'),
"%a %b %d %H:%M:%S +0000 %Y")
# get the time of the last tweet
if len(hashtag_dict.keys()) == 0:
time_last_tweet = created_at
else:
time_last_tweet = sorted(hashtag_dict.keys())[-1]
# ensure to only address tweets not over 60 seconds older
# than the last
time_delta = created_at - time_last_tweet
if time_delta.total_seconds() > - 60:
# if 'entities' in tweet_json:
if 'hashtags' in tweet_json['entities']:
hashtags = tweet_json['entities']['hashtags']
hashtag_list = [i['text'].encode('utf-8').strip()
for i in hashtags]
# keep only unique tags
hashtag_list = list(set(hashtag_list))
# update hashtag_dict with tweet of at least 2 tags
if len(hashtag_list) > 1:
hashtag_dict[created_at] = hashtag_list
# for those with 0 or 1 tag, the result is the
# same as in the last
else:
output.write(str(result) + '\n')
continue
else:
output.write(str(result) + '\n')
continue
# ignore tweets over 60 seconds older than the last
else:
continue
# remove the tweets more than 60 second older than the
# current tweet
sorted_key_list = sorted(hashtag_dict.keys())
for k in sorted_key_list:
time_elapsed = created_at - k
if time_elapsed.total_seconds() > 60:
del hashtag_dict[k]
else:
break
# finally, initialize the graph, and derive average degree
graph = Graph()
for t in hashtag_dict:
for tag in hashtag_dict[t]:
graph.add_vertex(tag)
length = len(hashtag_dict[t])
for start in range(length - 1):
for end in range(start+1, length):
graph.add_edge(
hashtag_dict[t][start],
hashtag_dict[t][end])
result = average_degree_graph(graph)
# ignore those lines without 'created_at'
else:
continue
# for each valid line, report the result
print result
output.write(str(result) + '\n')
else:
print 'All done!'
except IOError as err:
print 'File error:', str(err)
