def contiguous_verses():
assert passage.init(
[reference.init('Genesis', 1, 2),
reference.init('Genesis', 1, 1)]) == [
passage.Data('Genesis 1:1-2', [
reference.init('Genesis', 1, 1),
reference.init('Genesis', 1, 2)
], None)
]
def describe_tf_idf():
blank = None
topic_1_id = 'topic_1_id'
topic_2_id = 'topic_2_id'
passage1 = passage.init([reference.init('Genesis', 1, 2)])[0]
passage2 = passage.init([reference.init('Exodus', 5, 1)])[0]
passage3 = passage.init([reference.init('John', 3, 16)])[0]
sample = [
cluster_result.ClusterResult(0.5, 'sd281x', blank, blank,
[passage1]),
cluster_result.ClusterResult(0.25, 'ds432a', blank, blank,
[passage2, passage3])
]
passage_to_topic_ids = {
passage1.name: [topic_1_id],
passage2.name: [topic_1_id, topic_2_id],
passage3.name: [topic_2_id]
}
def to_tf_idf_topics():
result = cluster_result.to_tf_idf_topics(sample,
passage_to_topic_ids)
assert result == [[(topic_1_id, 1 * math.log(3 / 2))],
[(topic_1_id, 1 * math.log(3 / 2)),
(topic_2_id, 2 * math.log(3 / 2))]]
def to_mean_weighted_tf_idf_topics():
result = cluster_result.to_mean_weighted_tf_idf_topics(
sample, passage_to_topic_ids)
mean = lambda values: sum(values) / float(len(values))
print(result)
assert result == [
(topic_2_id, 0.25 * 2 * math.log(3 / 2)),
(topic_1_id,
mean([0.5 * 1 * math.log(3 / 2), 0.25 * 1 * math.log(3 / 2)]))
]
def duplicates():
assert passage.init([
reference.init('Genesis', 1, 2),
reference.init('Genesis', 1, 2),
reference.init('Genesis', 1, 1)
]) == [
passage.Data('Genesis 1:1-2', [
reference.init('Genesis', 1, 1),
reference.init('Genesis', 1, 2)
], None)
]
def mixed_verses():
assert passage.init([
reference.init('Genesis', 1, 5),
reference.init('Genesis', 1, 2),
reference.init('Genesis', 1, 1)
]) == [
passage.Data('Genesis 1:1-2,5', [
reference.init('Genesis', 1, 1),
reference.init('Genesis', 1, 2),
reference.init('Genesis', 1, 5)
], None)
]
def multiple_chapters():
assert passage.init([
reference.init('Genesis', 1, 5),
reference.init('Genesis', 1, 2),
reference.init('Genesis', 3, 12),
reference.init('Genesis', 3, 10),
reference.init('Genesis', 1, 1)
]) == [
passage.Data('Genesis 1:1-2,5', [
reference.init('Genesis', 1, 1),
reference.init('Genesis', 1, 2),
reference.init('Genesis', 1, 5)
], None),
passage.Data('Genesis 3:10,12', [
reference.init('Genesis', 3, 10),
reference.init('Genesis', 3, 12)
], None),
]
def multiple_books():
assert passage.init([
reference.init('Genesis', 1, 5),
reference.init('Genesis', 1, 2),
reference.init('Exodus', 5, 3),
reference.init('Genesis', 3, 12),
reference.init('Revelation', 1, 1),
reference.init('Genesis', 3, 10),
reference.init('Genesis', 1, 1)
]) == [
passage.Data('Genesis 1:1-2,5', [
reference.init('Genesis', 1, 1),
reference.init('Genesis', 1, 2),
reference.init('Genesis', 1, 5)
], None),
passage.Data('Genesis 3:10,12', [
reference.init('Genesis', 3, 10),
reference.init('Genesis', 3, 12)
], None),
passage.Data('Exodus 5:3', [reference.init('Exodus', 5, 3)], None),
passage.Data('Revelation 1:1',
[reference.init('Revelation', 1, 1)], None)
]
def by_topic():
attr = flip(getattr)
by_topic = compose(lambda v: v['topic_name'], second)
by_category = compose(lambda v: v['category_text'], second)
by_subtopic = compose(lambda v: v['subtopic_text'], second)
def _to_references(value):
return dissoc(
second(value), 'reference_list', 'source_topic_key',
'sort_order_category', 'subtopic_key', 'category_key', 'topic_key',
'sort_order_subtopic')
return {
topic: {
category: {
subtopic: passage.text(passage.init(list_map(first, values)))
for subtopic, values in groupby(by_subtopic,
chap_values).items()
}
for category, chap_values in groupby(by_category,
topic_values).items()
}
for topic, topic_values in groupby(by_topic, init()).items()
}
def single():
p = passage.init([reference.Data('Genesis', 1, 1)])[0]
frame = nave.topics_frame(p)
assert frame.label.iloc[0] == 'TIME'
assert len(frame) == 7