def test_init_from_sqlite3_conn_specific_cols(self):
from medspacy.io.db_connect import DbConnect
import sqlite3
sq_conn = sqlite3.connect(db)
cursor = sq_conn.cursor()
db_conn = DbConnect(conn=sq_conn)
cols = ["label_", "text", "is_negated", "section_category"]
db_dtypes = [
"varchar(100)",
"varchar(100)",
"int",
"varchar(100)",
]
from medspacy.io.db_writer import DbWriter
writer = DbWriter(db_conn,
"ents",
cols=cols,
col_types=db_dtypes,
create_table=True,
drop_existing=True)
writer.write(doc)
cursor.execute("SELECT * FROM ents;")
rslts = cursor.fetchone()
assert rslts == ("CONDITION", "pneumonia", True, None)
db_conn.close()
def test_init_from_sqlite3_conn(self):
from medspacy.io.db_connect import DbConnect
import sqlite3
sq_conn = sqlite3.connect(db)
db_conn = DbConnect(conn=sq_conn)
assert db_conn is not None
db_conn.close()
def test_init_from_sqlite3_conn_defaults(self):
"""Test writing with default values for ent attributes."""
sq_conn = sqlite3.connect(db)
cursor = sq_conn.cursor()
db_conn = DbConnect(conn=sq_conn)
from medspacy.io.db_writer import DbWriter
writer = DbWriter(db_conn, "ents", cols=None, col_types=None,
create_table=True, drop_existing=False)
writer.write(doc)
cursor.execute("SELECT * FROM ents;")
rslts = cursor.fetchone()
assert rslts == ('pneumonia', 24, 33, 'CONDITION', 1, 0, 0, 0, 0, None, None)
db_conn.close()
def test_init_from_sqlite3_conn(self):
from medspacy.io.db_connect import DbConnect
import sqlite3
create_test_db(db)
sq_conn = sqlite3.connect(db)
db_conn = DbConnect(conn=sq_conn)
from medspacy.io.db_reader import DbReader
reader = DbReader(db_conn, "SELECT text_id, text FROM texts")
rslts = reader.read()
assert rslts[0] == (1, "Patient with a history of breast ca")
assert rslts[1] == (2, "There is no evidence of pneumonia.")
db_conn.close()
def test_init_from_sqlite3_conn(self):
from medspacy.io.db_connect import DbConnect
import sqlite3
create_test_db(db)
sq_conn = sqlite3.connect(db)
db_conn = DbConnect(conn=sq_conn)
from medspacy.io.db_reader import DbReader
reader = DbReader(db_conn, "SELECT text_id, text FROM texts")
from medspacy.io.db_writer import DbWriter
writer = DbWriter(db_conn,
"ents",
["text_id"] + doc_consumer.dtype_attrs["ent"],
["int"] + db_dtypes,
create_table=True,
drop_existing=False)
from medspacy.io.pipeline import Pipeline
pipeline = Pipeline(reader, writer, nlp, "ent")
pipeline.process()
sq_conn = sqlite3.connect(db)
cursor = sq_conn.cursor()
cursor.execute("SELECT * FROM ents;")
rslts = cursor.fetchall()
assert rslts[0] == (1, "pneumonia", "CONDITION", True, None)
assert rslts[1] == (2, "breast ca", "CONDITION", False, None)
sq_conn.close()
"pattern": [{
"LOWER": "=",
"OP": "+"
}]
}])
consumer = DocConsumer(
nlp, context=True,
sectionizer=True) # DocConsumer has optional bool context and sectionizer
nlp.add_pipe(sectionizer)
nlp.add_pipe(context)
nlp.add_pipe(consumer)
#####################################################
#####################################################
# CREATING DB CONNECTIONS
db_read_conn = DbConnect(driver, server, db, user, pwd)
db_write_conn = DbConnect(driver, server, db, user, pwd)
#####################################################
#####################################################
# CREATING DB READ/WRITERS
db_reader = DbReader(db_read_conn, read_query, start, end,
read_batch_size) # start, end, read_batch_size optional
db_writer = DbWriter(db_write_conn, destination_table, create_table,
drop_existing, write_batch_size, cols, col_types)
#####################################################
#####################################################
# CREATING PIPELINE
pipeline = Pipeline(db_reader, db_writer, nlp, row_type)
#####################################################
def test_init_from_conn_info(self):
from medspacy.io.db_connect import DbConnect
db_conn = DbConnect(driver, server, db, user, pwd)
assert db_conn is not None