def get_recommendations(query, fold_no):
df_graph=None
source=query.split('\n')
i=len(source)
while not df_graph:
src_lines=source[:i]
if src_lines:
df_graph=ASTBuilder('\n'.join(source[:i])).build_AST()
i-=1
query_obj_types=[]
calls=[]
query_line=re.split(r'[^\w]',source[-1])
query_obj=filter(None, query_line)[-1]
assign_nodes, call_nodes= df_graph.find_assignments_and_calls(query_obj)
for node in assign_nodes:
query_obj_types.extend(node.src)
for node in call_nodes:
calls.append(node.tgt)
query_count=Counter(calls)
conn=sqlite3.connect("pyty.db")
c=conn.cursor()
objects=[]
for type in query_obj_types:
results=c.execute('''SELECT obj_calls FROM TRAINSET_{fold} WHERE obj_type=?'''.
format(fold=fold_no),(type,))
for obj in results:
obj_calls=obj[0].split(',')
score=compute_manhattan_dist(query_count,
Counter(obj_calls))
objects.append((obj_calls, score))
objects=sorted(objects, key=lambda tup: tup[0])
call_set=Counter()
for object in objects:
call_set.update(Counter(object[0])-query_count)
recommendations=[call[0] for call in call_set.most_common(10)]
return recommendations
def get_recommendations(query, fold_no):
df_graph = None
source = query.split('\n')
i = len(source)
while not df_graph:
src_lines = source[:i]
if src_lines:
df_graph = ASTBuilder('\n'.join(source[:i])).build_AST()
i -= 1
query_obj_types = []
calls = []
query_line = re.split(r'[^\w]', source[-1])
query_obj = filter(None, query_line)[-1]
assign_nodes, call_nodes = df_graph.find_assignments_and_calls(query_obj)
for node in assign_nodes:
query_obj_types.extend(node.src)
for node in call_nodes:
calls.append(node.tgt)
query_count = Counter(calls)
conn = sqlite3.connect("pyty.db")
c = conn.cursor()
objects = []
for type in query_obj_types:
results = c.execute(
'''SELECT obj_calls FROM TRAINSET_{fold} WHERE obj_type=?'''.
format(fold=fold_no), (type, ))
for obj in results:
obj_calls = obj[0].split(',')
score = compute_manhattan_dist(query_count, Counter(obj_calls))
objects.append((obj_calls, score))
objects = sorted(objects, key=lambda tup: tup[0])
call_set = Counter()
for object in objects:
call_set.update(Counter(object[0]) - query_count)
recommendations = [call[0] for call in call_set.most_common(10)]
return recommendations
