def test_benchmark_dbpedia_lookup_subject_columns_only():
onlyfiles = [
f for f in listdir(ENTITIES_DIR) if isfile(join(ENTITIES_DIR, f))
]
scidentifier = SCIdentifier()
num = 0
while True:
try:
_id = onlyfiles[num]
print("process table %d out of %d" % (num, len(onlyfiles)),
flush=True)
print("table id %s" % (_id), flush=True)
fixture_entities = get_gold_standard_entities(_id)
_table = GenericTable(filename=join(TABLES_DIR, _id), _id=_id)
_table.init()
_subject_columns = scidentifier.identify_subject_column(_table)
if _subject_columns:
_table.subject_column = _subject_columns[0]
dbpedia_lookup_entities = disambiguate_table_subject_column_only(
_table)
to_compare = map_agdistis_entities_to_gold_standard_format(
_table, dbpedia_lookup_entities)
print("", flush=True)
print(fixture_entities, flush=True)
print("", flush=True)
print(to_compare, flush=True)
print(diff_entities(fixture_entities, to_compare), flush=True)
num += 1
if (num >= len(onlyfiles)):
break
except BaseException as e:
print(str(e))
def test_disambiguate_table_subject_column_only_case_1():
table = GenericTable()
table.table = CASE_1_TABLE
table.subject_column = 1
entities = disambiguate_table_subject_column_only(table)
import ipdb
ipdb.set_trace()
def get_additional_tables(self):
tables = []
subject_column_list = os.path.join(ADDITIONAL_DATA_DIR,
"subject_columns.csv")
id_list = self.load_csv(subject_column_list)
for (_id, subject_column) in id_list:
table_filename = os.path.join(ADDITIONAL_DATA_DIR, "tables", _id)
table = GenericTable(filename=table_filename, _id=_id)
table.init()
table.table = table.table[:int(ROWS_TO_ANALYZE)]
table.subject_column = int(subject_column)
tables.append(table)
return tables