def setUp(self):
self.db = db
self.db.drop_all()
self.db.create_all()
seed_db(db)
AlchemyExtractor.API_KEY = 'fake'
self.ex = AlchemyExtractor()
self.doc = Document()
self.doc.text = 'foo'
def setUp(self):
self.db = db
self.db.drop_all()
self.db.create_all()
seed_db(db)
AlchemyExtractor.API_KEY = 'fake'
self.ex = AlchemyExtractor()
self.doc = Document()
self.doc.text = 'foo'
class TestAlchemyExtractor(unittest.TestCase):
def setUp(self):
self.db = db
self.db.drop_all()
self.db.create_all()
seed_db(db)
AlchemyExtractor.API_KEY = 'fake'
self.ex = AlchemyExtractor()
self.doc = Document()
self.doc.text = 'foo'
def tearDown(self):
self.db.session.remove()
self.db.drop_all()
def test_extract_entities(self):
entities = [
{
"type": "Person",
"relevance": "0.703385",
"count": "4",
"text": "Adam Welkom"
},
{
"type": "Person",
"relevance": "0.562693",
"count": "4",
"text": "Joyce Moamogwa",
"quotations": [{
"quotation": "\"We are not safe, we do not trust them. They are like our enemies,\" she said ...",
}]
}]
self.ex.extract_entities(self.doc, entities)
e = self.doc.entities[0]
self.assertEqual('Adam Welkom', e.entity.name)
self.assertEqual('person', e.entity.group)
self.assertEqual(4, e.count)
self.assertEqual(0.703385, e.relevance)
e = self.doc.entities[1]
self.assertEqual('Joyce Moamogwa', e.entity.name)
self.assertEqual('person', e.entity.group)
self.assertEqual(4, e.count)
self.assertEqual(0.562693, e.relevance)
def test_extract_keywords(self):
keywords = [
{
"relevance": "0.703385",
"text": "morning",
},
{
"relevance": "0.562693",
"text": "justice",
}]
self.ex.extract_keywords(self.doc, keywords)
kw = self.doc.keywords[0]
self.assertEqual('morning', kw.keyword)
self.assertEqual(0.703385, kw.relevance)
kw = self.doc.keywords[1]
self.assertEqual('justice', kw.keyword)
self.assertEqual(0.562693, kw.relevance)
def test_keyword_offsets(self):
self.doc.text = 'Oh what a beautiful morning for mob justice. In the morning. With justice.'
keywords = [
{
"relevance": "0.703385",
"text": "morning",
},
{
"relevance": "0.562693",
"text": "justice",
}]
self.ex.extract_keywords(self.doc, keywords)
k = self.doc.keywords[0]
self.assertEqual('morning', k.keyword)
self.assertEqual('20:7 52:7', k.offset_list)
k = self.doc.keywords[1]
self.assertEqual('justice', k.keyword)
self.assertEqual('36:7 66:7', k.offset_list)
def test_entity_offsets(self):
self.doc.text = 'Foo Adam Welkom: "We are not safe," she said. Another Adam Welkom.'
entities = [
{
"type": "Person",
"relevance": "0.703385",
"count": "4",
"text": "Adam Welkom"
},
]
self.ex.extract_entities(self.doc, entities)
e = self.doc.entities[0]
self.assertEqual('Adam Welkom', e.entity.name)
self.assertEqual('4:11 54:11', e.offset_list)
def test_all_offsets(self):
offsets = self.ex.all_offsets('foo bar baz bar bam', 'bar')
self.assertEqual('4:3 12:3', offsets)
def test_utterance_offsets(self):
self.doc.text = 'Foo Adam Welkom: "We are not safe," she said. Another Adam Welkom.'
entities = [
{
"type": "Person",
"relevance": "0.562693",
"count": "4",
"text": "Joyce Moamogwa",
"quotations": [{
"quotation": "\"We are not safe,\" she said ...",
}]
}]
self.ex.extract_entities(self.doc, entities)
u = self.doc.utterances[0]
self.assertEqual(17, u.offset)
self.assertEqual(27, u.length)
class TestAlchemyExtractor(unittest.TestCase):
def setUp(self):
self.db = db
self.db.drop_all()
self.db.create_all()
seed_db(db)
AlchemyExtractor.API_KEY = 'fake'
self.ex = AlchemyExtractor()
self.doc = Document()
self.doc.text = 'foo'
def tearDown(self):
self.db.session.remove()
self.db.drop_all()
def test_extract_entities(self):
entities = [
{
"type": "Person",
"relevance": "0.703385",
"count": "4",
"text": "Adam Welkom"
},
{
"type": "Person",
"relevance": "0.562693",
"count": "4",
"text": "Joyce Moamogwa",
"quotations": [{
"quotation": "\"We are not safe, we do not trust them. They are like our enemies,\" she said ...",
}]
}]
self.ex.extract_entities(self.doc, entities)
e = self.doc.entities[0]
self.assertEqual('Adam Welkom', e.entity.name)
self.assertEqual('person', e.entity.group)
self.assertEqual(4, e.count)
self.assertEqual(0.703385, e.relevance)
e = self.doc.entities[1]
self.assertEqual('Joyce Moamogwa', e.entity.name)
self.assertEqual('person', e.entity.group)
self.assertEqual(4, e.count)
self.assertEqual(0.562693, e.relevance)
def test_extract_keywords(self):
keywords = [
{
"relevance": "0.703385",
"text": u"morning",
},
{
"relevance": "0.562693",
"text": u"justice",
}]
self.ex.extract_keywords(self.doc, keywords)
kw = self.doc.keywords[0]
self.assertEqual('morning', kw.keyword)
self.assertEqual(0.703385, kw.relevance)
kw = self.doc.keywords[1]
self.assertEqual('justice', kw.keyword)
self.assertEqual(0.562693, kw.relevance)
def test_keyword_offsets(self):
self.doc.text = 'Oh what a beautiful morning for mob justice. In the morning. With justice.'
keywords = [
{
"relevance": "0.703385",
"text": u"morning",
},
{
"relevance": "0.562693",
"text": u"justice",
}]
self.ex.extract_keywords(self.doc, keywords)
k = self.doc.keywords[0]
self.assertEqual('morning', k.keyword)
self.assertEqual('20:7 52:7', k.offset_list)
k = self.doc.keywords[1]
self.assertEqual('justice', k.keyword)
self.assertEqual('36:7 66:7', k.offset_list)
def test_entity_offsets(self):
self.doc.text = 'Foo Adam Welkom: "We are not safe," she said. Another Adam Welkom.'
entities = [
{
"type": "Person",
"relevance": "0.703385",
"count": "4",
"text": "Adam Welkom"
},
]
self.ex.extract_entities(self.doc, entities)
e = self.doc.entities[0]
self.assertEqual('Adam Welkom', e.entity.name)
self.assertEqual('4:11 54:11', e.offset_list)
def test_all_offsets(self):
offsets = self.ex.all_offsets('foo bar baz bar bam', 'bar')
self.assertEqual('4:3 12:3', offsets)
def test_utterance_offsets(self):
self.doc.text = 'Foo Adam Welkom: "We are not safe," she said. Another Adam Welkom.'
entities = [
{
"type": "Person",
"relevance": "0.562693",
"count": "4",
"text": "Joyce Moamogwa",
"quotations": [{
"quotation": "\"We are not safe,\" she said ...",
}]
}]
self.ex.extract_entities(self.doc, entities)
u = self.doc.utterances[0]
self.assertEqual(17, u.offset)
self.assertEqual(27, u.length)
def test_extract_taxonomy(self):
taxonomy = [{
"label": "/law, govt and politics/law enforcement/police",
"score": "0.538782"
}, {
"confident": "no",
"label": "/law, govt and politics",
"score": "0.380401"
}, {
"confident": "no",
"label": "/technology and computing/internet technology/social network",
"score": "0.135708"
}]
self.ex.extract_taxonomy(self.doc, taxonomy)
tx = self.doc.taxonomies[0]
self.assertEqual('/law, govt and politics/law enforcement/police', tx.label)
self.assertEqual(0.538782, tx.score)
self.assertEqual(1, len(self.doc.taxonomies))
def test_extract_taxonomy_no_confident(self):
taxonomy = [{
"confident": "no",
"label": "/law, govt and politics/law enforcement/police",
"score": "0.538782"
}, {
"confident": "no",
"label": "/law, govt and politics",
"score": "0.380401"
}, {
"confident": "no",
"label": "/technology and computing/internet technology/social network",
"score": "0.135708"
}]
self.ex.extract_taxonomy(self.doc, taxonomy)
tx = self.doc.taxonomies[0]
self.assertEqual('/law, govt and politics/law enforcement/police', tx.label)
self.assertEqual(0.538782, tx.score)
self.assertEqual(3, len(self.doc.taxonomies))
class TestAlchemyExtractor(unittest.TestCase):
def setUp(self):
self.db = db
self.db.drop_all()
self.db.create_all()
seed_db(db)
AlchemyExtractor.API_KEY = 'fake'
self.ex = AlchemyExtractor()
self.doc = Document()
self.doc.text = 'foo'
def tearDown(self):
self.db.session.remove()
self.db.drop_all()
def test_extract_entities(self):
entities = [{
"type": "Person",
"relevance": "0.703385",
"count": "4",
"text": "Adam Welkom"
}, {
"type":
"Person",
"relevance":
"0.562693",
"count":
"4",
"text":
"Joyce Moamogwa",
"quotations": [{
"quotation":
"\"We are not safe, we do not trust them. They are like our enemies,\" she said ...",
}]
}]
self.ex.extract_entities(self.doc, entities)
e = self.doc.entities[0]
self.assertEqual('Adam Welkom', e.entity.name)
self.assertEqual('person', e.entity.group)
self.assertEqual(4, e.count)
self.assertEqual(0.703385, e.relevance)
e = self.doc.entities[1]
self.assertEqual('Joyce Moamogwa', e.entity.name)
self.assertEqual('person', e.entity.group)
self.assertEqual(4, e.count)
self.assertEqual(0.562693, e.relevance)
def test_extract_keywords(self):
keywords = [{
"relevance": "0.703385",
"text": u"morning",
}, {
"relevance": "0.562693",
"text": u"justice",
}]
self.ex.extract_keywords(self.doc, keywords)
kw = self.doc.keywords[0]
self.assertEqual('morning', kw.keyword)
self.assertEqual(0.703385, kw.relevance)
kw = self.doc.keywords[1]
self.assertEqual('justice', kw.keyword)
self.assertEqual(0.562693, kw.relevance)
def test_keyword_offsets(self):
self.doc.text = 'Oh what a beautiful morning for mob justice. In the morning. With justice.'
keywords = [{
"relevance": "0.703385",
"text": u"morning",
}, {
"relevance": "0.562693",
"text": u"justice",
}]
self.ex.extract_keywords(self.doc, keywords)
k = self.doc.keywords[0]
self.assertEqual('morning', k.keyword)
self.assertEqual('20:7 52:7', k.offset_list)
k = self.doc.keywords[1]
self.assertEqual('justice', k.keyword)
self.assertEqual('36:7 66:7', k.offset_list)
def test_entity_offsets(self):
self.doc.text = 'Foo Adam Welkom: "We are not safe," she said. Another Adam Welkom.'
entities = [
{
"type": "Person",
"relevance": "0.703385",
"count": "4",
"text": "Adam Welkom"
},
]
self.ex.extract_entities(self.doc, entities)
e = self.doc.entities[0]
self.assertEqual('Adam Welkom', e.entity.name)
self.assertEqual('4:11 54:11', e.offset_list)
def test_all_offsets(self):
offsets = self.ex.all_offsets('foo bar baz bar bam', 'bar')
self.assertEqual('4:3 12:3', offsets)
def test_utterance_offsets(self):
self.doc.text = 'Foo Adam Welkom: "We are not safe," she said. Another Adam Welkom.'
entities = [{
"type":
"Person",
"relevance":
"0.562693",
"count":
"4",
"text":
"Joyce Moamogwa",
"quotations": [{
"quotation": "\"We are not safe,\" she said ...",
}]
}]
self.ex.extract_entities(self.doc, entities)
u = self.doc.utterances[0]
self.assertEqual(17, u.offset)
self.assertEqual(27, u.length)
def test_extract_taxonomy(self):
taxonomy = [{
"label": "/law, govt and politics/law enforcement/police",
"score": "0.538782"
}, {
"confident": "no",
"label": "/law, govt and politics",
"score": "0.380401"
}, {
"confident": "no",
"label":
"/technology and computing/internet technology/social network",
"score": "0.135708"
}]
self.ex.extract_taxonomy(self.doc, taxonomy)
tx = self.doc.taxonomies[0]
self.assertEqual('/law, govt and politics/law enforcement/police',
tx.label)
self.assertEqual(0.538782, tx.score)
self.assertEqual(1, len(self.doc.taxonomies))
def test_extract_taxonomy_no_confident(self):
taxonomy = [{
"confident": "no",
"label": "/law, govt and politics/law enforcement/police",
"score": "0.538782"
}, {
"confident": "no",
"label": "/law, govt and politics",
"score": "0.380401"
}, {
"confident": "no",
"label":
"/technology and computing/internet technology/social network",
"score": "0.135708"
}]
self.ex.extract_taxonomy(self.doc, taxonomy)
tx = self.doc.taxonomies[0]
self.assertEqual('/law, govt and politics/law enforcement/police',
tx.label)
self.assertEqual(0.538782, tx.score)
self.assertEqual(3, len(self.doc.taxonomies))
