def test_chexpert_extractor():
extractor = RegExExtractor(
__tests_dir / 'data/patterns/chexpert_phrases.yml', 'CheXpert labeler')
dir = get_example_dir()
with open(dir / '1.chexpert.xml') as fp:
c = bioc.load(fp)
actual_documents = c.documents
expected_documents = []
for doc in actual_documents:
doc = copy.deepcopy(doc)
for p in doc.passages:
del p.annotations[:]
expected_documents.append(doc)
for expected_doc, actual_doc in zip(expected_documents, actual_documents):
extractor.__call__(expected_doc)
expected_anns = sorted(list(
bioc.annotations(expected_doc, bioc.PASSAGE)),
key=lambda a: a.total_span.offset)
actual_anns = sorted(list(bioc.annotations(actual_doc, bioc.PASSAGE)),
key=lambda a: a.total_span.offset)
assert len(expected_anns) == len(actual_anns), \
'{} vs {}'.format(len(expected_anns), len(actual_anns))
for expected_ann, actual_ann in zip(expected_anns, actual_anns):
assert expected_ann.total_span == actual_ann.total_span
for k in ['observation', 'annotator']:
assert expected_ann.infons[k] == actual_ann.infons[k]
def test_annotations():
results = list(bioc.annotations(collection))
assert {'1', '2', '3', '4', '5'} == {r.annotation.id for r in results}
results = list(bioc.annotations(collection, level=DOCUMENT))
assert {'5'} == {r.annotation.id for r in results}
assert {'1'} == {r.document.id for r in results}
assert {None} == {r.passage for r in results}
results = list(bioc.annotations(collection, level=PASSAGE))
assert {'1', '2'} == {r.annotation.id for r in results}
assert {'1'} == {r.document.id for r in results}
assert {0} == {r.passage.offset for r in results}
assert {None} == {r.sentence for r in results}
results = list(bioc.annotations(collection, level=SENTENCE))
assert {'3', '4'} == {r.annotation.id for r in results}
assert {'2'} == {r.document.id for r in results}
assert {27} == {r.passage.offset for r in results}
assert {27, 34} == {r.sentence.offset for r in results}
results = list(bioc.annotations(collection.documents[0], level=SENTENCE))
assert len(results) == 0
with pytest.raises(TypeError):
next(bioc.annotations('Foo'))
def test_sibling_intext_citations(table_article):
all_passages = []
all_annotations = []
file = StringIO(table_article)
for doc in docs2bioc(file, 'pmcxml', trim_sentences=False):
all_passages.extend(doc.passages)
all_annotations.extend(bioc.annotations(doc))
assert any([
'inspected using the graphics program PyMOL.' in chunk.text
for chunk in all_passages
])
assert '[14],[16],[23]\u2013[25]' in [
a.infons['citation_text'] for a in all_annotations
]
def test_annotations():
annotations = list(bioc.annotations(collection))
assert 2 == len(annotations)
assert '1' == annotations[0].id
assert '2' == annotations[1].id
annotations = list(bioc.annotations(collection, level=bioc.SENTENCE))
assert 2 == len(annotations)
assert '3' == annotations[0].id
assert '4' == annotations[1].id
annotations = list(bioc.annotations(collection, level=bioc.DOCUMENT))
assert 1 == len(annotations)
assert '5' == annotations[0].id
annotations = list(
bioc.annotations(collection.documents[1], level=bioc.SENTENCE))
assert 2 == len(annotations)
assert '4' == annotations[1].id
with pytest.raises(TypeError):
next(bioc.annotations('Foo'))
with pytest.raises(ValueError):
next(bioc.annotations(collection, level=-1))
with pytest.raises(ValueError):
next(
bioc.annotations(collection.documents[0].passages[0],
level=bioc.DOCUMENT))
with pytest.raises(ValueError):
next(
bioc.annotations(collection.documents[1].passages[0].sentences[0],
level=bioc.DOCUMENT))
next(
bioc.annotations(collection.documents[1].passages[0].sentences[0],
level=bioc.PASSAGE))
def __extract_tags(self, tagged_article, document, wanted_tags):
"""
This function extract the words corresponding to the wanted_tags and
lemmatizes these, to reduce redundancy. These words are saved in
the Tag object.
:param tagged_article: A tagged article.
:param document: The document corresponding to the tagged_article.
:param wanted_tags: The words corresponding to these tags are saved.
:return: An Tag object containing the article information + the wanted words.
"""
for anno in list(bioc.annotations(document)):
tag = anno.infons['type']
for wanted_tag in wanted_tags:
if wanted_tag == tag:
word = LEMMATIZER.lemmatize(
anno.text
) # e.g. rats --> rat, so to reduce redundancy
tagged_article.add_annotation(tag, word)
return tagged_article