def get(self, term):
"""
Returns list of matching concepts or entities using lexical search
"""
args = simple_parser.parse_args()
q = GolrSearchQuery(term, user_agent=USER_AGENT, **args)
results = q.search()
return results
def test_search_go_ontol():
q = GolrSearchQuery("transport*", category='ontology_class', is_go=True)
print("Q={}".format(q))
params = q.solr_params()
print("PARAMS={}".format(params))
results = q.search()
print("RESULTS={}".format(results))
docs = results.docs
for doc in docs:
print(str(doc))
assert len(docs) > 0
def test_search():
q = GolrSearchQuery("abnormal")
print("Q={}".format(q))
params = q.solr_params()
print("PARAMS={}".format(params))
results = q.search()
print("RESULTS={}".format(results))
docs = results.docs
for r in docs:
print(str(r))
assert len(docs) > 0
def test_search_go_all():
q = GolrSearchQuery("transport*", is_go=True)
print("Q={}".format(q))
params = q.solr_params()
print("PARAMS={}".format(params))
results = q.search()
print("RESULTS={}".format(results))
docs = results.docs
for r in docs:
print(str(r))
assert len(docs) > 0
print(str(results.facet_counts))
def test_cursor():
"""
Tests rows and start parameters.
First fetch 100 docs, then same query but iterate with cursor in increments of ten.
The two sets of IDs returned should be identical
"""
q = GolrSearchQuery("abnormal", rows=100)
results = q.search()
docs = results.docs
ids = set([d['id'] for d in docs])
print('Init ids={}'.format(ids))
assert len(ids) == 100
matches = set()
for i in range(0, 10):
q = GolrSearchQuery("abnormal", start=i * 10, rows=10)
docs = q.search().docs
next_ids = [d['id'] for d in docs]
assert len(next_ids) == 10
print('Next ids (from {}) = {}'.format(i * 10, next_ids))
matches.update(next_ids)
assert len(matches) == 100
assert len(matches.intersection(ids)) == 100
def clean_feature_ids(id: str) -> str:
"""
MME queries often need to be sanitized before going into owlsim, for example:
MIM:610536 -> OMIM:610536
SHH -> HGNC:10848
"""
if ':' in id:
prefix, reference = id.split(':')
if prefix == 'MIM':
id = 'OMIM:' + reference
else:
# Assume it's a label and look it up
# Assume it's human, and make sure it's an exact match
query = GolrSearchQuery(id, taxon=['NCBITaxon:9606'], min_match="100%")
results = query.search()
if results.docs:
id = results.docs[0]['id']
return id
class TestGolrSearchQuery():
@classmethod
def setup_class(self):
self.manager = GolrSearchQuery()
# Mock the PySolr search function to
# return our test docs
input_fh = os.path.join(os.path.dirname(__file__),
'resources/solr/solr-docs.json')
input_docs = json.load(open(input_fh))
self.test_results = pysolr.Results(input_docs)
self.manager.solr.search = MagicMock(return_value=self.test_results)
@classmethod
def teardown_class(self):
self.manager = None
def test_longest_hl(self):
test_data = [
"<em>Muscle</em> <em>atrophy</em>, generalized",
"Generalized <em>muscle</em> degeneration",
"Diffuse skeletal <em>muscle</em> wasting"
]
expected = "<em>Muscle</em> <em>atrophy</em>, generalized"
results = self.manager._get_longest_hl(test_data)
assert expected == results
def test_longest_hl_ambiguous(self):
test_data = [
"<em>Muscle</em> <em>atrophy</em>, generalized",
"Generalized <em>muscle</em> degeneration",
"Diffuse skeletal <em>muscle</em> wasting",
"<em>Muscle</em> <em>atrophy</em>, not generalized",
]
expected = "<em>Muscle</em> <em>atrophy</em>, generalized"
results = self.manager._get_longest_hl(test_data)
assert expected == results
def test_hl_to_string(self):
test_data = "Foo <em>Muscle</em> bar <em>atrophy</em>, generalized"
expected = "Foo Muscle bar atrophy, generalized"
results = self.manager._hl_as_string(test_data)
assert expected == results
def test_invalid_xml(self):
test_data = "Foo<Foo> <em>Muscle</em> bar <em>atrophy</em>, generalized"
pytest.raises(ET.ParseError, self.manager._hl_as_string, test_data)
def test_autocomplete_doc_conversion(self):
"""
Given a sample solr output as a pysolr.Results object
test that _process_autocomplete_results returns the
expected object
"""
expected_fh = os.path.join(os.path.dirname(__file__),
'resources/solr/autocomplete-expected.json')
processed_docs = json.load(open(expected_fh))
output_docs = self.manager._process_autocomplete_results(self.test_results)
assert json.dumps(processed_docs, sort_keys=True) == \
json.dumps(output_docs,
default=lambda obj: getattr(obj, '__dict__', str(obj)),
sort_keys=True)
def test_search_doc_conversion(self):
"""
Given a sample solr output as a pysolr.Results object
test that _process_autocomplete_results returns the
expected object
"""
expected_fh = os.path.join(os.path.dirname(__file__),
'resources/solr/search-expected.json')
processed_docs = json.load(open(expected_fh))
output_docs = self.manager._process_search_results(self.test_results)
assert json.dumps(processed_docs, sort_keys=True) == \
json.dumps(output_docs,
default=lambda obj: getattr(obj, '__dict__', str(obj)),
sort_keys=True)
def test_search(self):
"""
Given a mock PySolr.search method test that
search() returns the expected object
"""
expected_fh = os.path.join(os.path.dirname(__file__),
'resources/solr/search-expected.json')
processed_docs = json.load(open(expected_fh))
output_docs = self.manager.search()
assert json.dumps(processed_docs, sort_keys=True) == \
json.dumps(output_docs,
default=lambda obj: getattr(obj, '__dict__', str(obj)),
sort_keys=True)
def test_autocomplete(self):
"""
Given a mock PySolr.search method test that
autocomplete() returns the expected object
"""
expected_fh = os.path.join(os.path.dirname(__file__),
'resources/solr/autocomplete-expected.json')
processed_docs = json.load(open(expected_fh))
output_docs = self.manager.autocomplete()
assert json.dumps(processed_docs, sort_keys=True) == \
json.dumps(output_docs,
default=lambda obj: getattr(obj, '__dict__', str(obj)),
sort_keys=True)
def test_autocomplete_no_category(self):
"""
Test for document without a category
"""
# Provide a new mock file
input_fh = os.path.join(os.path.dirname(__file__),
'resources/solr/autocomplete-nocat.json')
input_docs = json.load(open(input_fh))
self.test_results = pysolr.Results(input_docs)
self.manager.solr.search = MagicMock(return_value=self.test_results)
expected_fh = os.path.join(os.path.dirname(__file__),
'resources/solr/autocomplete-nocat-expect.json')
processed_docs = json.load(open(expected_fh))
output_docs = self.manager.autocomplete()
print(json.dumps(output_docs,
default=lambda obj: getattr(obj, '__dict__', str(obj)),
sort_keys=True))
assert json.dumps(processed_docs, sort_keys=True) == \
json.dumps(output_docs,
default=lambda obj: getattr(obj, '__dict__', str(obj)),
sort_keys=True)
def search(term, args):
q = GolrSearchQuery(term, args)
return q.search()
