def test_text_case_sensitivity():
text = 'hello Hello HELLO'
annotator = Annotator(
'Hello',
patterns=('hello', ),
)
annotations = list(annotator.annotate(text))
assert annotations == [
Annotation('hello', (0, 5), type='Hello'),
Annotation('Hello', (6, 11), type='Hello'),
Annotation('HELLO', (12, 17), type='Hello'),
]
annotator = Annotator(
'AlternativeHello',
patterns=('Hello', ),
settings={'case_sensitive': True},
)
annotations = list(annotator.annotate(text))
assert annotations == [
Annotation('Hello', (6, 11), type='AlternativeHello'),
]
def test_eq():
A = Annotation('hello', (0, 5), type='Greeting', score=1.0)
B = Annotation('hello', (0, 5), type='Greeting', score=1.0)
C = Annotation('world', (0, 5), type='Place', score=1.0)
assert A == B
assert A is not B
assert A != C
def test_ordering():
A = Annotation('hello', (0, 5))
B = Annotation('hell', (0, 4))
C = Annotation('world', (6, 11))
annotations = AnnotationList([C, A])
annotations.append(B)
assert annotations.index(A) == 0
assert annotations.index(B) == 1
assert annotations.index(C) == 2
def test_disambiguate():
A1 = Annotation('hello', (0, 5), score=1.1)
A2 = Annotation('hell', (0, 4), score=1.0)
B = Annotation('world', (6, 11), score=1.1)
annotations = AnnotationList([A1, A2, B])
assert annotations.disambiguate() == (A1, B)
assert A2 in annotations
annotations.disambiguate(discard_others=True)
assert A2 not in annotations
def test_cells():
A1 = Annotation('hello', (0, 5), type='Greeting')
A2 = Annotation('hell', (0, 4), type='Place')
B1 = Annotation('world!', (6, 12), type='Place')
B2 = Annotation('world', (6, 11), type='Place')
C = Annotation('!', (11, 12), type='Symbol')
annotations = AnnotationList([A1, A2, B1, B2, C])
assert annotations.cells == [
[A1, A2],
[B1, B2],
[C],
]
def test_regex_case_sensitivity():
text = 'HELLO Bello cello'
annotator = Annotator(
'Thing',
patterns='/[hbc]ello/',
)
annotations = list(annotator.annotate(text))
assert annotations == [
Annotation('HELLO', (0, 5), type='Thing'),
Annotation('Bello', (6, 11), type='Thing'),
Annotation('cello', (12, 17), type='Thing'),
]
def test_representation():
text = 'hello world'
annotator = Annotator(
'HelloWorld',
patterns={
'hello': 'Hello',
'world': 'World',
},
)
annotations = list(annotator.annotate(text))
assert annotations == [
Annotation('hello', (0, 5), type='HelloWorld', data='Hello'),
Annotation('world', (6, 11), type='HelloWorld', data='World'),
]
def test_agent():
text = '''
hello
world
'''
agent = Agent()
agent.add_annotator(Annotator(
'HelloWorld',
patterns=('hello', 'world'),
))
annotated_text = agent.annotate(text)
assert len(annotated_text.lines) == 4
assert annotated_text.lines[1].cells[0][0] == \
Annotation('hello', (0, 5), type='HelloWorld')
assert annotated_text.lines[2].cells[0][0] == \
Annotation('world', (0, 5), type='HelloWorld')
def test_agent_create_annotator():
text = 'hello world'
agent = Agent()
agent.create_annotator(
'Hello',
patterns='hello',
)
annotated_text = agent.annotate(text)
assert annotated_text.lines[0].annotations == [
Annotation('hello', (0, 5), type='Hello'),
]
def test_filter_and_boost():
A1 = Annotation('hello', (0, 5), type='Greeting', score=1.0)
A2 = Annotation('hell', (0, 4), type='Place', score=1.0)
B = Annotation('world', (6, 11), type='Place', score=1.0)
annotations = AnnotationList([A1, A2, B])
assert annotations.filter(type='Greeting') == [A1]
assert annotations.filter(type='Place') == [A2, B]
annotations.filter(type='Greeting').boost(1.25)
assert A1.score == 1.25
assert A2.score == 1.0
assert B.score == 1.0
annotations.cells[0].filter(type='Place').boost(1.50)
assert A1.score == 1.25
assert A2.score == 1.50
assert B.score == 1.0
annotations.cells[1].boost(1.75)
assert A1.score == 1.25
assert A2.score == 1.50
assert B.score == 1.75
def test_line_add(A, B, L1, L2, T1, T2):
T3 = T1 + T2
# verify integrity of source objects
assert A.span == (0, 5)
assert B.span == (0, 5)
assert T1.annotations == L1 == [A]
assert T2.annotations == L2 == [B]
assert T1.text == 'hello'
assert T2.text == 'world'
# verify validity of created object
B2 = Annotation('world', (6, 11))
assert T3.text == 'hello world'
assert T3.annotations == [A, B2]
def test_tokens():
text = 'hello, world!'
annotator = Annotator(
'Thing',
tokens={
'comma': ',',
'dot': '.',
'sep': '/({comma}|{dot})? /',
},
patterns='hello{sep}world!',
)
annotations = list(annotator.annotate(text))
assert annotations == [
Annotation('hello, world!', (0, 13), type='Thing'),
]
def test_combinations():
A1 = Annotation('hello', (0, 5))
A2 = Annotation('hell', (0, 4))
A3 = Annotation('hello world', (0, 11))
B1 = Annotation('world', (6, 11))
B2 = Annotation('worl', (6, 10))
B3 = Annotation('world!', (6, 12))
C = Annotation('!', (11, 12))
annotations = AnnotationList([A1, A2, A3, B1, B2, B3, C])
assert set(annotations.combinations()) == {
(A1, B1, C),
(A1, B2, C),
(A1, B3),
(A2, B1, C),
(A2, B2, C),
(A2, B3),
(A3, C),
}
def A():
return Annotation('hello', (0, 5))
def B():
return Annotation('world', (0, 5))