class DBTestCase(unittest.TestCase):
def setUp(self):
self.db = DBHelper()
self.samples =self.db.get_results_by_clms(columns='*',table=
'samples',database=DB_test,isdict=True)
def tearDown(self):
self.db = None
def test_get_max_group_id(self):
res =self.db.get_max_group_id('refined_info_2','xyzb')
res_2 =self.db.get_max_group_id('info_template','xyzb')
print res
print res_2
def test_batch_insert(self):
entrys =self.db.get_results_by_clms(columns='*',table=
'refined_info',database=DB_1,isdict=True)[:12056]
table = 'info_tmp1'
self.db.recr_table(table,DB_1,'info_template',DB_1);
self.db.batch_insert(entrys,table,DB_1)
def test_batch_get(self):
entrys =self.db.get_results_by_clms(columns='*',table=
'refined_info',database=DB_1,isdict=True)[:2]
for key, value in entrys[1].items():
print key,value,type(value)
def test_diff_tables(self):
dc = datachecker()
db_helper = DBHelper()
new_table = 'extracted_full_info_today'
old_table = 'info_tmp'
old_db = DB_test
new_db = DB_1
db_helper.recr_table(old_table,old_db,'info_template',DB_1)
entrys =dc.diff_tables(table_old=old_table,table_new=new_table,db_old=old_db,db_new=new_db)
db_helper.batch_insert(entrys,table=old_table,db=old_db)
class scheduler():
""" To schedule the data flow
"""
def __init__(self):
self.db = DBHelper()
self.dc = datachecker()
self.sb = submitter()
self.dd = data_drawer()
self.sg = segger()
self.id = indexer()
self.pc = prechecker()
self.dmg_on = DMG_ON
self.test_on = TEST_ON
if len(sys.argv) < 4:
self.table_old = "refined_info"
self.db_old = DB_1
self.table_new = "extracted_info"
self.db_new = DB_1
else:
self.table_old = sys.argv[1]
self.db_old = sys.argv[2]
self.table_new = sys.argv[3]
self.db_new = sys.argv[4]
def data_flow(self):
threads = []
self.pc.pre_process()
#remove duplicates
entrys = self.dc.diff_tables(self.table_old,self.table_new,self.db_old,self.db_new)
#insert the new coming entrys into database
self.db.batch_insert(entrys,self.table_old,self.db_old)
if (self.test_on):
t = threading.Thread(target=self.db.batch_insert,args=(entrys,self.table_old,DB_test))
t.start()
threads.append(t)
if(self.dmg_on):
#submite the updates to dmg
self.sb.deal_submit_req(entrys)
#get the all data from dmg
entrys = dd.get_entrys()
self.db.recr_table(TB_DMG,DB_1,TB_TEMPLATE,DB_1)
t = threading.Thread(target=self.db.batch_insert,args=(entrys,TB_DMG,DB_1))
t.start()
threads.append(t)
else:
entrys = self.db.get_results_by_clms(columns='*',table= self.table_old,
database=self.db_old,isdict=True)
entrys = sorted(entrys,key=self.sort_func,reverse=True)
#serilize entrys to protobuffers
entrys = self.sg.serilize_entrys(entrys)
#remote call word segging
entrys = self.sg.word_seg_list(entrys)
# #remote call indexing
self.id.index_list_string(entrys)
# wait until all threads are over
for t in threads:
t.join()
def sort_func(self,entry):
info_type = entry.get(INFO_TYPE)
if info_type == 1:
time = entry.get(MEETING_TIME)
else:
time = entry.get(RELEASE_DATE)
if time is None:
return 0
else :
return get_timestamp(time)
