def test_sentence_graph(self):
s_1 = Sentence([
Clause(self.word_1, self.word_2, self.word_10),
Clause(self.word_1, self.word_3, self.word_10),
Clause(self.word_1, self.word_4, self.word_10),
Clause(self.word_3, self.word_9, self.word_10),
Clause(self.word_3, self.word_5, self.word_10),
Clause(self.word_4, self.word_6, self.word_10),
Clause(self.word_5, self.word_8, self.word_10),
Clause(self.word_6, self.word_7, self.word_10)
])
s_2 = Sentence([
Clause(self.word_2, self.word_1, self.word_10),
Clause(self.word_2, self.word_5, self.word_10),
Clause(self.word_2, self.word_3, self.word_10),
Clause(self.word_1, self.word_4, self.word_10),
Clause(self.word_5, self.word_7, self.word_10),
Clause(self.word_5, self.word_8, self.word_10),
Clause(self.word_3, self.word_9, self.word_10),
Clause(self.word_4, self.word_6, self.word_10)
])
intersection = set(s_1.tree_graph.nodes) & set(s_1.tree_graph.nodes)
graph = build_graph(s_1, s_2, intersection)
correct_graph = {
self.word_1: [self.word_2, self.word_4],
self.word_2: [self.word_1],
self.word_3: [self.word_9],
self.word_4: [self.word_1, self.word_6],
self.word_5: [self.word_8],
self.word_6: [self.word_4],
self.word_7: [],
self.word_8: [self.word_5],
self.word_9: [self.word_3],
}
# The node adjacency list should be sorted
for v_1, v_2 in zip(correct_graph.values(), graph.values()):
v_1.sort()
v_2.sort()
pp = pprint.PrettyPrinter()
print("Built graph: ")
pp.pprint(graph)
print("Correct graph: ")
pp.pprint(correct_graph)
self.assertEqual(graph, correct_graph)
def test_partition_sentences(self):
s_1 = Sentence([
Clause(self.word_1, self.word_2, self.word_10),
Clause(self.word_1, self.word_3, self.word_10),
Clause(self.word_1, self.word_4, self.word_10),
Clause(self.word_3, self.word_9, self.word_10),
Clause(self.word_3, self.word_5, self.word_10),
Clause(self.word_4, self.word_6, self.word_10),
Clause(self.word_5, self.word_8, self.word_10),
Clause(self.word_6, self.word_7, self.word_10)
])
s_2 = Sentence([
Clause(self.word_2, self.word_1, self.word_10),
Clause(self.word_2, self.word_5, self.word_10),
Clause(self.word_2, self.word_3, self.word_10),
Clause(self.word_1, self.word_4, self.word_10),
Clause(self.word_5, self.word_7, self.word_10),
Clause(self.word_5, self.word_8, self.word_10),
Clause(self.word_3, self.word_9, self.word_10),
Clause(self.word_4, self.word_6, self.word_10)
])
intersection = list(
set(s_1.tree_graph.nodes) & set(s_2.tree_graph.nodes))
intersection.sort()
graph = build_graph(s_1, s_2, intersection)
partitions = partition_graph(graph)
correct_partitions = [{
self.word_1, self.word_2, self.word_4, self.word_6
}, {self.word_3, self.word_9}, {self.word_5, self.word_8},
{self.word_7}]
# This is because the order doesn't matter to us but the partition_graph method returns a list
def same_partitions(p1, p2):
if len(p1) != len(p2):
return False
else:
return all(partition in p2 for partition in p1)
self.assertTrue(same_partitions(partitions, correct_partitions))
def test_grammatical_class_index_sentences(self):
s_1 = Sentence([
Clause(self.word_1, self.word_2, self.word_5),
Clause(self.word_1, self.word_4, self.word_7),
Clause(self.word_1, self.word_6, self.word_9),
Clause(self.word_2, self.word_3, self.word_7),
Clause(self.word_2, self.word_8, self.word_5),
Clause(self.word_6, self.word_10, self.word_5)
])
s_2 = Sentence([
Clause(self.word_4, self.word_1, self.word_7),
Clause(self.word_4, self.word_6, self.word_8),
Clause(self.word_1, self.word_3, self.word_9),
Clause(self.word_1, self.word_10, self.word_2),
Clause(self.word_6, self.word_5, self.word_9)
])
s_3 = Sentence([
Clause(self.word_9, self.word_2, self.word_1),
Clause(self.word_2, self.word_6, self.word_3),
Clause(self.word_2, self.word_4, self.word_3),
Clause(self.word_2, self.word_8, self.word_7),
Clause(self.word_4, self.word_10, self.word_7)
])
s_4 = Sentence([
Clause(self.word_8, self.word_7, self.word_1),
Clause(self.word_7, self.word_6, self.word_2),
Clause(self.word_6, self.word_4, self.word_3),
Clause(self.word_6, self.word_5, self.word_9)
])
s_5 = Sentence([
Clause(self.word_8, self.word_7, self.word_4),
Clause(self.word_8, self.word_10, self.word_3)
])
s_6 = Sentence([
Clause(self.word_6, self.word_3, self.word_1),
Clause(self.word_6, self.word_4, self.word_10),
Clause(self.word_4, self.word_7, self.word_9)
])
usl_a = usl(Text([s_1, s_2, s_6, s_5]))
usl_b = usl(Text([s_2, s_3, s_6, s_4]))
eo_index = grammatical_class_index(usl_a, usl_b, 'EO', Sentence)
oo_index = grammatical_class_index(usl_a, usl_b, 'OO', Sentence)
print("EO index: " + str(eo_index))
print("OO index: " + str(oo_index))
self.assertTrue(eo_index != 1,
"Two different USLs don't have an EO index of 1")
self.assertTrue(oo_index != 1,
"Two different USLs don't have an EO index of 1")
def test_connexity_index_sentence(self):
s_1 = Sentence([
Clause(self.word_1, self.word_2, self.word_5),
Clause(self.word_1, self.word_4, self.word_7),
Clause(self.word_1, self.word_6, self.word_9),
Clause(self.word_2, self.word_3, self.word_7),
Clause(self.word_2, self.word_8, self.word_5),
Clause(self.word_6, self.word_10, self.word_5)
])
s_2 = Sentence([
Clause(self.word_4, self.word_1, self.word_7),
Clause(self.word_4, self.word_6, self.word_8),
Clause(self.word_1, self.word_3, self.word_9),
Clause(self.word_1, self.word_10, self.word_2),
Clause(self.word_6, self.word_5, self.word_9)
])
s_3 = Sentence([
Clause(self.word_9, self.word_2, self.word_1),
Clause(self.word_2, self.word_6, self.word_3),
Clause(self.word_2, self.word_4, self.word_3),
Clause(self.word_2, self.word_8, self.word_7),
Clause(self.word_4, self.word_10, self.word_7)
])
s_4 = Sentence([
Clause(self.word_8, self.word_7, self.word_1),
Clause(self.word_7, self.word_6, self.word_2),
Clause(self.word_6, self.word_4, self.word_3),
Clause(self.word_6, self.word_5, self.word_9)
])
s_5 = Sentence([
Clause(self.word_8, self.word_7, self.word_4),
Clause(self.word_8, self.word_10, self.word_3)
])
s_6 = Sentence([
Clause(self.word_6, self.word_3, self.word_1),
Clause(self.word_6, self.word_4, self.word_10),
Clause(self.word_4, self.word_7, self.word_9)
])
usl_a = usl(Text([s_1, s_2, s_6, s_5]))
usl_b = usl(Text([s_2, s_3, s_6, s_4]))
index = connexity_index(Sentence, usl_a, usl_b)
print("Proximity Index for the different USLs: " + str(index))
self.assertTrue(
index != 1 and index != 0,
"Different USLs should yield and index that isn't null nor is 1")
index = connexity_index(Sentence, usl_a, usl_a)
print("Proximity Index for the same USLs: " + str(index))
self.assertTrue(index == 1,
"Identical USLs should yield and index of 1")
def test_super_sentence_graph(self):
s_1 = Sentence([
Clause(self.word_1, self.word_2, self.word_5),
Clause(self.word_1, self.word_4, self.word_7),
Clause(self.word_1, self.word_6, self.word_9),
Clause(self.word_2, self.word_3, self.word_7),
Clause(self.word_2, self.word_8, self.word_5),
Clause(self.word_6, self.word_10, self.word_5)
])
s_2 = Sentence([
Clause(self.word_4, self.word_1, self.word_7),
Clause(self.word_4, self.word_6, self.word_8),
Clause(self.word_1, self.word_3, self.word_9),
Clause(self.word_1, self.word_10, self.word_2),
Clause(self.word_6, self.word_5, self.word_9)
])
s_3 = Sentence([
Clause(self.word_9, self.word_2, self.word_1),
Clause(self.word_2, self.word_6, self.word_3),
Clause(self.word_2, self.word_4, self.word_3),
Clause(self.word_2, self.word_8, self.word_7),
Clause(self.word_4, self.word_10, self.word_7)
])
s_4 = Sentence([
Clause(self.word_8, self.word_7, self.word_1),
Clause(self.word_7, self.word_6, self.word_2),
Clause(self.word_6, self.word_4, self.word_3),
Clause(self.word_6, self.word_5, self.word_9)
])
s_5 = Sentence([
Clause(self.word_8, self.word_7, self.word_4),
Clause(self.word_8, self.word_10, self.word_3)
])
s_6 = Sentence([
Clause(self.word_6, self.word_3, self.word_1),
Clause(self.word_6, self.word_4, self.word_10),
Clause(self.word_4, self.word_7, self.word_9)
])
super_sentence_1 = SuperSentence([
SuperClause(s_1, s_2, s_6),
SuperClause(s_1, s_3, s_6),
SuperClause(s_3, s_4, s_6)
])
super_sentence_2 = SuperSentence([
SuperClause(s_2, s_1, s_6),
SuperClause(s_2, s_3, s_6),
SuperClause(s_3, s_4, s_6)
])
graph = build_graph(
super_sentence_1, super_sentence_2,
set(super_sentence_1.tree_graph.nodes)
& set(super_sentence_2.tree_graph.nodes))
correct_graph = {s_1: [s_2], s_2: [s_1], s_3: [s_4], s_4: [s_3]}
# The node adjacency list should be sorted
for v_1, v_2 in zip(correct_graph.values(), graph.values()):
v_1.sort()
v_2.sort()
pp = pprint.PrettyPrinter()
print("Built graph: ")
pp.pprint(graph)
print("Correct graph: ")
pp.pprint(correct_graph)
self.assertEqual(graph, correct_graph)
def test_mutual_inclusion_super_sentence(self):
s_1 = Sentence([
Clause(self.word_1, self.word_2, self.word_5),
Clause(self.word_1, self.word_4, self.word_7),
Clause(self.word_1, self.word_6, self.word_9),
Clause(self.word_2, self.word_3, self.word_7),
Clause(self.word_2, self.word_8, self.word_5),
Clause(self.word_6, self.word_10, self.word_5)
])
s_2 = Sentence([
Clause(self.word_4, self.word_1, self.word_7),
Clause(self.word_4, self.word_6, self.word_8),
Clause(self.word_1, self.word_3, self.word_9),
Clause(self.word_1, self.word_10, self.word_2),
Clause(self.word_6, self.word_5, self.word_9)
])
s_3 = Sentence([
Clause(self.word_9, self.word_2, self.word_1),
Clause(self.word_2, self.word_6, self.word_3),
Clause(self.word_2, self.word_4, self.word_3),
Clause(self.word_2, self.word_8, self.word_7),
Clause(self.word_4, self.word_10, self.word_7)
])
s_4 = Sentence([
Clause(self.word_8, self.word_7, self.word_1),
Clause(self.word_7, self.word_6, self.word_2),
Clause(self.word_6, self.word_4, self.word_3),
Clause(self.word_6, self.word_5, self.word_9)
])
s_5 = Sentence([
Clause(self.word_8, self.word_7, self.word_4),
Clause(self.word_8, self.word_10, self.word_3)
])
s_6 = Sentence([
Clause(self.word_6, self.word_3, self.word_1),
Clause(self.word_6, self.word_4, self.word_10),
Clause(self.word_4, self.word_7, self.word_9)
])
super_sentence_1 = SuperSentence(
[SuperClause(s_1, s_2, s_3),
SuperClause(s_1, s_6, s_4)])
super_sentence_2 = SuperSentence([
SuperClause(s_4, s_2, s_5),
SuperClause(s_4, s_1, s_6),
SuperClause(s_4, s_3, s_5)
])
super_sentence_3 = SuperSentence([
SuperClause(s_6, s_1, s_3),
SuperClause(s_1, s_2, s_4),
SuperClause(s_2, s_5, s_3)
])
super_sentence_4 = SuperSentence([
SuperClause(s_4, s_2, s_6),
SuperClause(s_4, s_1, s_6),
SuperClause(s_2, s_3, s_6)
])
usl_a = usl(
Text([super_sentence_1, super_sentence_2, super_sentence_3]))
usl_b = usl(
Text([super_sentence_1, super_sentence_2, super_sentence_4]))
index = mutual_inclusion_index(usl_a, usl_b)
print("Proximity Index for the different USLs: " + str(index))
self.assertTrue(
index != 1 and index != 0,
"Different USLs should yield and index that isn't null nor is 1")
index = mutual_inclusion_index(usl_a, usl_a)
print("Proximity Index for the identical USLs: " + str(index))
self.assertTrue(index == 1,
"Identical USLs should yield and index of 1")