def strong_components(self):
"""
Generates sets containing the sets of nodes in each strongly
connected component (U{http://mathworld.wolfram.com/StronglyConnectedComponent.html}).
@rtype: C{generator}
@return: a generator of sets of nodes for each strongly connected component
"""
import traversals
visited = set()
order = list(traversals.dfs(self, post=True))
top_order = reversed(order)
for node in top_order:
if not node in visited:
component = set()
for node in traversals.dfs(self, node, pre=True, adj=self.parents):
if not node in visited:
component.add(node)
visited.add(node)
yield component
def flatten(self):
# flatten the structure (in place) so that
# all nodes are either leaf or parents of leaf nodes
for node in dfs(self, post=True):
grand_children = set()
for child in list(self.children[node]):
if self.outdegree(child):
self.del_edge((node, child))
grand_children |= self.children[child]
for gc in grand_children:
self.add_edge((node, gc))
def flatten(self):
# flatten the structure (in place) so that
# all nodes are either leaf or parents of leaf nodes
for node in dfs(self, post=True):
grand_children = set()
for child in list(self.children[node]):
if self.outdegree(child):
self.del_edge((node, child))
grand_children |= self.children[child]
for gc in grand_children:
self.add_edge((node, gc))
def structure_is_valid(self):
# the induced graph containing the nodes reachable from
# each root should be a tree
for root in self.root_nodes:
visited = set()
num_edges = 0
num_nodes = 0
for node in dfs(self, root, post=True):
num_nodes += 1
num_edges += self.outdegree(node)
if not num_nodes == num_edges + 1:
return False
return True
def structure_is_valid(self):
# the induced graph containing the nodes reachable from
# each root should be a tree
for root in self.root_nodes:
visited = set()
num_edges = 0
num_nodes = 0
for node in dfs(self, root, post=True):
num_nodes += 1
num_edges += self.outdegree(node)
if not num_nodes == num_edges + 1:
return False
return True
def deepen(self):
# deepen structure (in place) to provide
# nicer layout for humans
for node in list(dfs(self, post=True)):
children = list(self.children[node])
if children:
it = iter(children)
new_parents = set(self.parents[next(it)])
new_parents.remove(node)
for child in it:
if not new_parents:
break
new_parents &= self.parents[child]
for p in new_parents:
for child in children:
self.del_edge((p, child))
self.add_edge((p, node))
def deepen(self):
# deepen structure (in place) to provide
# nicer layout for humans
for node in list(dfs(self, post=True)):
children = list(self.children[node])
if children:
it = iter(children)
new_parents = set(self.parents[next(it)])
new_parents.remove(node)
for child in it:
if not new_parents:
break
new_parents &= self.parents[child]
for p in new_parents:
for child in children:
self.del_edge((p, child))
self.add_edge((p, node))
def reachable(self, node):
# return a generator of the leaf
# nodes reachable from node
for n in dfs(self, node, post=True):
if not self.children[n]:
yield n
def reachable(self, node):
# return a generator of the leaf
# nodes reachable from node
for n in dfs(self, node, post=True):
if not self.children[n]:
yield n
from content import distances, city_adjacency
from traversals import dfs, bfs, iter_deep_dfs, bidir_search, \
greedily_best_first_search, min_sum_estimation, informational_search
from decision_tree import visualize, make_dot_file
start = "Харьков"
end = "Ниж.Новгород"
print('__Неинформированный поиск__')
# DFS
dfs_dist, dfs_way = dfs(city_adjacency, start, end)
print(f'DFS:\t\t\t\t\t\t{dfs_dist, dfs_way}')
# BFS
bfs_dist, bfs_way = bfs(city_adjacency, start, end)
print(f'BFS:\t\t\t\t\t\t{bfs_dist, bfs_way}')
# с ограничением глубины:
max_depth = 7
md_dfs_dist, md_dfs_way = dfs(city_adjacency, start, end, max_depth=max_depth)
print(f'DFS with max_depth of {max_depth}:\t{md_dfs_dist, md_dfs_way}')
# с итеративным углублением
id_dfs_dist, id_dfs_way = iter_deep_dfs(city_adjacency, start, end)
print(f'Iteration deepening:\t\t{id_dfs_dist, id_dfs_way}')
# с двунаправленным поиском
bds_dist, bds_way = bidir_search(city_adjacency, start, end)
print(f'Bidirectional search:\t\t{bds_dist, bds_way}')
print('__Информативный поиск__')
# жадный поиск по первому наилучшему соответствию
gbfs_dist, gbfs_way = informational_search(greedily_best_first_search,
city_adjacency, distances, start,