def clear():
connection = db_tools.connect()
tables = ['Users', 'Forums', 'Threads', 'Posts', 'Follow', 'Subscribe']
db_tools.execute(connection, "SET global foreign_key_checks = 1;")
for table in tables:
query = "TRUNCATE TABLE %s;" % table
db_tools.execute(connection, query)
db_tools.execute(connection, "SET global foreign_key_checks = 1;")
connection.close()
return jsonify({"code": STATUS_CODE['OK'], "response": 'OK'})
def clear():
connection = db_tools.connect()
tables = ['Users', 'Forums', 'Threads', 'Posts', 'Follow', 'Subscribe']
db_tools.execute(connection, "SET global foreign_key_checks = 1;")
for table in tables:
query = "TRUNCATE TABLE %s;" % table
db_tools.execute(connection, query)
db_tools.execute(connection, "SET global foreign_key_checks = 1;")
connection.close()
return jsonify({"code": STATUS_CODE['OK'], "response": 'OK'})
def get_data():
x_feats = 7
y_feats = 1
connection = db_tools.connect()
cur = connection.cursor()
# get number of rows so training array can be built
cur.execute("SELECT count(id) FROM answers")
data = cur.fetchone()
num_rows = data[0]
build_X = np.ndarray(shape=(num_rows,x_feats), dtype=float, order='C')
build_y = np.ndarray(shape=(num_rows,), dtype=float, order='C')
# get all data necessary to train
cur.execute("SELECT body,score,commcount,bodysentipolarity,bodysentisubjectivity, links, promptness, acceptedanswer FROM answers")
x_indx = 0
y_indx = 0
while True:
data = cur.fetchone()
if data == None:
break
body = data[0]
bodylen = len(body)
score = data[1]
commcount = data[2]
sentipolarity = data[3]
sentisubjectivity = data[4]
links = data[5]
promptness = data[6]
acceptedanswer = data[7]
if acceptedanswer == None:
aa = 0
else:
aa = 1
np.put(build_X, [range(x_indx,x_indx + x_feats)], [bodylen,score,commcount,sentipolarity,sentisubjectivity,links,promptness]) #bodylen,score,commcount,sentipolarity,sentisubjectivity,links,promptness
np.put(build_y, [range(y_indx,y_indx + y_feats)], [aa])
x_indx += x_feats
y_indx += y_feats
return build_X, build_y
def get_data():
connection = db_tools.connect()
cur = connection.cursor()
q_group = question_group()
master_map = defaultdict(int)
# get all data necessary to train
cur.execute("SELECT body,score,commcount,bodysentipolarity,bodysentisubjectivity,links,promptness,acceptedanswer,pid,id FROM answers")
while True:
data = cur.fetchone()
if data == None:
break
q_id = data[8]
if master_map[q_id] == 0:
master_map[q_id] = [q_group.build_object(data)]
else:
master_map[q_id].append(q_group.build_object(data))
#print master_map
#print "\n\n\n\n\n\n"
return master_map
elem.clear()
while elem.getprevious() is not None:
del elem.getparent()[0]
if __name__=='__main__':
# DATASET MUST BE IN SOURCE FOLDER WITHIN A FOLDER CALLED 'datasets'
static_path = os.getcwd()
users = static_path+'/../datasets/Users.xml'
posts = static_path+'/../datasets/Posts.xml'
# connect to the dataset
context = etree.iterparse(posts)
# connect to the database
print("Connecting to DB")
connection = db_tools.connect()
print("Setting up DB (if necessary)")
db_tools.setup_db(connection)
print("Populating DB")
fast_iter(context, connection, limit=10000)
connection.close()
del context
#coding:utf-8
import sys
import string
import db_tools
from bson.son import SON
mydb=db_tools.connect()
items=mydb.service.find({})
for item in items:
print item['rating'].encode('utf-8')
'''
r=mydb.command(SON([('geoNear', 'service'), ('near', [116.34, 39.99]), ('spherical', True), ('num', 1000), ('distanceMultiplier', 6378137), ('maxDistance', 5000.0/6378137), ('query', {'category': '搬家'})]))
for item in r['results']:
obj=item['obj']
print obj['telephone']
'''