def __init__(self, dbinfo=ct.DB_INFO, redis_host=None):
self.dbinfo = dbinfo
self.logger = getLogger(__name__)
self.index_objs = dict()
self.stock_objs = dict()
self.updating_date = None
self.combination_objs = dict()
self.cal_client = CCalendar(dbinfo, redis_host)
self.index_info_client = IndexInfo()
self.cvaluation_client = CValuation()
self.reviewer = CReivew(dbinfo, redis_host)
self.comb_info_client = CombinationInfo(dbinfo, redis_host)
self.stock_info_client = CStockInfo(dbinfo, redis_host)
self.rindex_stock_data_client = RIndexStock(dbinfo, redis_host)
self.industry_info_client = IndustryInfo(dbinfo, redis_host)
self.rindustry_info_client = RIndexIndustryInfo(dbinfo, redis_host)
self.animation_client = CAnimation(dbinfo, redis_host)
self.subscriber = Subscriber()
self.quote_handler = StockQuoteHandler()
self.ticker_handler = TickerHandler()
self.connect_client = StockConnect(market_from=ct.SH_MARKET_SYMBOL,
market_to=ct.HK_MARKET_SYMBOL,
dbinfo=dbinfo,
redis_host=redis_host)
self.margin_client = Margin(dbinfo=dbinfo, redis_host=redis_host)
self.emotion_client = Emotion(dbinfo=dbinfo, redis_host=redis_host)
self.sh_exchange_client = StockExchange(ct.SH_MARKET_SYMBOL)
self.sz_exchange_client = StockExchange(ct.SZ_MARKET_SYMBOL)
def choose_plate(edate = '2016-10-11', ndays = 90):
rindustry_info_client = RIndexIndustryInfo(redis_host='127.0.0.1')
today_industry_df = rindustry_info_client.get_k_data(edate)
pchange_df = today_industry_df.sort_values(by = 'pchange', ascending = False).head(3)
mchange_df = today_industry_df.sort_values(by = 'mchange', ascending = False).head(3)
plate_code_list = list(set(pchange_df.code.tolist()).intersection(pchange_df.code.tolist()))
if len(plate_code_list) == 0:
logger.info("no interested plate for date:%s" % edate)
return list()
sdate = get_day_nday_ago(edate, ndays, '%Y-%m-%d')
#get sh index data
sh_index_obj = CIndex('000001', redis_host='127.0.0.1')
sh_index_info = sh_index_obj.get_k_data_in_range(sdate, edate)
sh_index_pchange = 100 * (sh_index_info.loc[len(sh_index_info) - 1, 'close'] - sh_index_info.loc[0, 'preclose']) / sh_index_info.loc[0, 'preclose']
#get industry data
all_industry_df = rindustry_info_client.get_k_data_in_range(sdate, edate)
all_industry_df = all_industry_df.loc[all_industry_df.code.isin(plate_code_list)]
industry_static_info = DataFrame(columns={'code', 'sai', 'pchange', ct.KL, ct.QL, ct.JL, ct.FL})
#choose better industry
redisobj = create_redis_obj("127.0.0.1")
today_industry_info = IndustryInfo.get(redisobj)
for code, industry in all_industry_df.groupby('code'):
industry = industry.reset_index(drop = True)
industry['sri'] = 0
industry['sri'] = industry['pchange'] - sh_index_info['pchange']
industry['sai'] = 0
industry.at[(industry.pchange > 0) & (sh_index_info.pchange < 0), 'sai'] = industry.loc[(industry.pchange > 0) & (sh_index_info.pchange < 0), 'sri']
industry_sai = len(industry.loc[industry.sai > 0])
industry_pchange = 100 * (industry.loc[len(industry) - 1, 'close'] - industry.loc[0, 'preclose']) / industry.loc[0, 'preclose']
code_list = json.loads(today_industry_info.loc[today_industry_info.code == code, 'content'].values[0])
info_dict, good_code_list = choose_stock(code_list, sdate, edate)
industry_static_info = industry_static_info.append(DataFrame([[code, industry_sai, industry_pchange, info_dict[ct.KL], info_dict[ct.QL], info_dict[ct.JL], info_dict[ct.FL]]], columns = ['code', 'sai', 'pchange', ct.KL, ct.QL, ct.JL, ct.FL]), sort = 'True')
industry_static_info = industry_static_info.reset_index(drop = True)
industry_static_info = industry_static_info.sort_values(by=['pchange'], ascending=False)
return good_code_list
def generate_data(self, cdate):
good_list = list()
obj_pool = Pool(500)
all_df = pd.DataFrame()
industry_info = IndustryInfo.get(self.redis)
failed_list = industry_info.code.tolist()
cfunc = partial(self.get_industry_data, cdate)
failed_count = 0
while len(failed_list) > 0:
is_failed = False
self.logger.debug("restart failed ip len(%s)" % len(failed_list))
for code_data in obj_pool.imap_unordered(cfunc, failed_list):
if code_data[1] is not None:
tem_df = code_data[1]
tem_df['code'] = code_data[0]
all_df = all_df.append(tem_df)
failed_list.remove(code_data[0])
else:
is_failed = True
if is_failed:
failed_count += 1
if failed_count > 10:
self.logger.info("%s rindustry init failed" % failed_list)
return pd.DataFrame()
time.sleep(10)
obj_pool.join(timeout=5)
obj_pool.kill()
self.mysql_client.changedb(self.get_dbname())
if all_df.empty: return all_df
all_df = all_df.reset_index(drop=True)
return all_df
def get_industry_data(self, cdate):
ri = RIndexIndustryInfo()
df = ri.get_k_data(cdate)
if df.empty: return df
df = df.reset_index(drop=True)
df = df.sort_values(by='amount', ascending=False)
df['money_change'] = (df['amount'] - df['preamount']) / 1e8
industry_info = IndustryInfo.get()
df = pd.merge(df, industry_info, how='left', on=['code'])
return df
def get_industry_data(self, _date):
df = pd.DataFrame()
df_info = IndustryInfo.get()
for _, code in df_info.code.iteritems():
data = CIndex(code).get_k_data(date=_date)
df = df.append(data)
df = df.reset_index(drop=True)
df['name'] = df_info['name']
df = df.sort_values(by='amount', ascending=False)
df = df.reset_index(drop=True)
return df
def __init__(self,
dbinfo=ct.DB_INFO,
redis_host=None,
stocks_dir='/data/tdx/history/days',
stock_path='/data/tdx/base/stocks.csv'):
self.table = ct.STOCK_INFO_TABLE
self.redis = create_redis_obj(
) if redis_host is None else create_redis_obj(host=redis_host)
self.mysql_client = cmysql.CMySQL(dbinfo, iredis=self.redis)
self.mysql_dbs = self.mysql_client.get_all_databases()
self.stocks_dir = stocks_dir
self.stock_path = stock_path
self.industry_info = IndustryInfo.get_industry()
class DataManager:
def __init__(self, dbinfo=ct.DB_INFO, redis_host=None):
self.dbinfo = dbinfo
self.logger = getLogger(__name__)
self.index_objs = dict()
self.stock_objs = dict()
self.updating_date = None
self.combination_objs = dict()
self.cal_client = CCalendar(dbinfo, redis_host)
self.index_info_client = IndexInfo()
self.cvaluation_client = CValuation()
self.reviewer = CReivew(dbinfo, redis_host)
self.comb_info_client = CombinationInfo(dbinfo, redis_host)
self.stock_info_client = CStockInfo(dbinfo, redis_host)
self.rindex_stock_data_client = RIndexStock(dbinfo, redis_host)
self.industry_info_client = IndustryInfo(dbinfo, redis_host)
self.rindustry_info_client = RIndexIndustryInfo(dbinfo, redis_host)
self.animation_client = CAnimation(dbinfo, redis_host)
self.subscriber = Subscriber()
self.quote_handler = StockQuoteHandler()
self.ticker_handler = TickerHandler()
self.connect_client = StockConnect(market_from=ct.SH_MARKET_SYMBOL,
market_to=ct.HK_MARKET_SYMBOL,
dbinfo=dbinfo,
redis_host=redis_host)
self.margin_client = Margin(dbinfo=dbinfo, redis_host=redis_host)
self.emotion_client = Emotion(dbinfo=dbinfo, redis_host=redis_host)
self.sh_exchange_client = StockExchange(ct.SH_MARKET_SYMBOL)
self.sz_exchange_client = StockExchange(ct.SZ_MARKET_SYMBOL)
def is_collecting_time(self):
now_time = datetime.now()
_date = now_time.strftime('%Y-%m-%d')
y, m, d = time.strptime(_date, "%Y-%m-%d")[0:3]
aft_open_hour, aft_open_minute, aft_open_second = (17, 10, 00)
aft_open_time = datetime(y, m, d, aft_open_hour, aft_open_minute,
aft_open_second)
aft_close_hour, aft_close_minute, aft_close_second = (23, 59, 59)
aft_close_time = datetime(y, m, d, aft_close_hour, aft_close_minute,
aft_close_second)
#self.logger.info("collecting now time. open_time:%s < now_time:%s < close_time:%s" % (aft_open_time, now_time, aft_close_time))
return aft_open_time < now_time < aft_close_time
def is_morning_time(self, now_time=datetime.now()):
_date = now_time.strftime('%Y-%m-%d')
y, m, d = time.strptime(_date, "%Y-%m-%d")[0:3]
mor_open_hour, mor_open_minute, mor_open_second = (0, 0, 0)
mor_open_time = datetime(y, m, d, mor_open_hour, mor_open_minute,
mor_open_second)
mor_close_hour, mor_close_minute, mor_close_second = (6, 30, 0)
mor_close_time = datetime(y, m, d, mor_close_hour, mor_close_minute,
mor_close_second)
return mor_open_time < now_time < mor_close_time
def collect_combination_runtime_data(self):
def _combination_run(code_id):
self.combination_objs[code_id].run()
return (code_id, True)
todo_iplist = list(self.combination_objs.keys())
return concurrent_run(_combination_run, todo_iplist, num=10)
def collect_stock_runtime_data(self):
if self.ticker_handler.empty(): return
datas = self.ticker_handler.getQueue()
while not datas.empty():
df = datas.get()
df = df.set_index('time')
df.index = pd.to_datetime(df.index)
for code_str in set(df.code):
code_id = code_str.split('.')[1]
self.stock_objs[code_id].run(df.loc[df.code == code_str])
def init_real_stock_info(self):
concerned_list = self.comb_info_client.get_concerned_list()
prefix_concerned_list = [add_prifix(code) for code in concerned_list]
ret = self.subscriber.subscribe(prefix_concerned_list, SubType.TICKER,
self.ticker_handler)
if 0 == ret:
for code in concerned_list:
if code not in self.stock_objs:
self.stock_objs[code] = CStock(code,
self.dbinfo,
should_create_influxdb=True,
should_create_mysqldb=True)
return ret
def init_index_info(self):
index_list = ct.INDEX_DICT.keys()
prefix_index_list = [add_index_prefix(code) for code in index_list]
ret = self.subscriber.subscribe(prefix_index_list, SubType.QUOTE,
self.quote_handler)
if 0 != ret:
self.logger.error("subscribe for index list failed")
return ret
for code in index_list:
if code not in self.index_objs:
self.index_objs[code] = CIndex(code,
should_create_influxdb=True,
should_create_mysqldb=True)
return 0
def collect_index_runtime_data(self):
if self.quote_handler.empty(): return
datas = self.quote_handler.getQueue()
while not datas.empty():
df = datas.get()
df['time'] = df.data_date + ' ' + df.data_time
df = df.drop(['data_date', 'data_time'], axis=1)
df = df.set_index('time')
df.index = pd.to_datetime(df.index)
for code_str in set(df.code):
code_id = code_str.split('.')[1]
self.index_objs[code_id].run(df.loc[df.code == code_str])
def run(self, sleep_time):
while True:
try:
self.logger.debug("enter run")
if self.cal_client.is_trading_day():
if is_trading_time():
t_sleep_time = 1
if not self.subscriber.status():
self.subscriber.start()
if 0 == self.init_index_info(
) and 0 == self.init_real_stock_info():
self.init_combination_info()
else:
self.logger.debug("enter stop subscriber")
self.subscriber.stop()
else:
self.collect_stock_runtime_data()
self.collect_combination_runtime_data()
self.collect_index_runtime_data()
self.animation_client.collect()
else:
t_sleep_time = sleep_time
if self.subscriber.status():
self.subscriber.stop()
else:
t_sleep_time = sleep_time
except Exception as e:
#traceback.print_exc()
self.logger.error(e)
gevent.sleep(t_sleep_time)
def set_update_info(self,
step_length,
exec_date,
cdate=None,
filename=ct.STEPFILE):
step_info = dict()
if cdate is None: cdate = 'none'
step_info[cdate] = dict()
step_info[cdate]['step'] = step_length
step_info[cdate]['date'] = exec_date
with open(filename, 'w') as f:
json.dump(step_info, f)
self.logger.info("finish step :%s" % step_length)
def get_update_info(self,
cdate=None,
exec_date=None,
filename=ct.STEPFILE):
if cdate is None: cdate = 'none'
if not os.path.exists(filename): return (0, exec_date)
with open(filename, 'r') as f:
step_info = json.load(f)
if cdate not in step_info: return (0, exec_date)
return (step_info[cdate]['step'], step_info[cdate]['date'])
def bootstrap(self,
cdate=None,
exec_date=datetime.now().strftime('%Y-%m-%d'),
ndays=3):
finished_step, exec_date = self.get_update_info(cdate, exec_date)
self.logger.info("enter updating.%s" % finished_step)
if finished_step < 1:
if not self.cal_client.init():
self.logger.error("cal client init failed")
return False
self.set_update_info(1, exec_date, cdate)
if finished_step < 2:
if not self.index_info_client.update():
self.logger.error("index info init failed")
return False
self.set_update_info(2, exec_date, cdate)
if finished_step < 3:
if not self.stock_info_client.update():
self.logger.error("stock info init failed")
return False
self.set_update_info(3, exec_date, cdate)
if finished_step < 4:
if not self.comb_info_client.update():
self.logger.error("comb info init failed")
return False
self.set_update_info(4, exec_date, cdate)
if finished_step < 5:
if not self.industry_info_client.update():
self.logger.error("industry info init failed")
return False
self.set_update_info(5, exec_date, cdate)
if finished_step < 6:
if not self.init_tdx_index_info(cdate):
self.logger.error("init tdx index info failed")
return False
self.set_update_info(6, exec_date, cdate)
if finished_step < 7:
if not self.sh_exchange_client.update(exec_date, num=ndays):
self.logger.error("sh exchange update failed")
return False
self.set_update_info(7, exec_date, cdate)
if finished_step < 8:
if not self.sz_exchange_client.update(exec_date, num=ndays):
self.logger.error("sz exchange update failed")
return False
self.set_update_info(8, exec_date, cdate)
if finished_step < 9:
if not self.init_index_components_info(exec_date):
self.logger.error("init index components info failed")
return False
self.set_update_info(9, exec_date, cdate)
if finished_step < 10:
if not self.init_industry_info(cdate):
self.logger.error("init industry info failed")
return False
self.set_update_info(10, exec_date, cdate)
if finished_step < 11:
if not self.rindustry_info_client.update(exec_date, num=ndays):
self.logger.error("init %s rindustry info failed" % exec_date)
return False
self.set_update_info(11, exec_date, cdate)
if finished_step < 12:
if not self.init_yesterday_hk_info(exec_date, num=ndays):
self.logger.error("init yesterday hk info failed")
return False
self.set_update_info(12, exec_date, cdate)
if finished_step < 13:
if not self.margin_client.update(exec_date, num=ndays):
self.logger.error("init yesterday margin failed")
return False
self.set_update_info(13, exec_date, cdate)
if finished_step < 14:
if not self.init_stock_info(cdate):
self.logger.error("init stock info set failed")
return False
self.set_update_info(14, exec_date, cdate)
if finished_step < 15:
if not self.init_base_float_profit():
self.logger.error("init base float profit for all stock")
return False
self.set_update_info(15, exec_date, cdate)
if finished_step < 16:
if not self.init_valuation_info(cdate):
self.logger.error("init stock valuation info failed")
return False
self.set_update_info(16, exec_date, cdate)
if finished_step < 17:
if not self.init_rvaluation_info(cdate):
self.logger.error("init r stock valuation info failed")
return False
self.set_update_info(17, exec_date, cdate)
if finished_step < 18:
if not self.init_rindex_valuation_info(cdate):
self.logger.error("init r index valuation info failed")
return False
self.set_update_info(18, exec_date, cdate)
if finished_step < 19:
if not self.rindex_stock_data_client.update(exec_date, num=ndays):
self.logger.error("rstock data set failed")
return False
self.set_update_info(19, exec_date, cdate)
if finished_step < 20:
if not self.set_bull_stock_ratio(exec_date, num=ndays):
self.logger.error("bull ratio set failed")
return False
self.set_update_info(20, exec_date, cdate)
self.logger.info("updating succeed")
return True
def clear_network_env(self):
kill_process("google-chrome")
kill_process("renderer")
kill_process("Xvfb")
kill_process("zygote")
kill_process("defunct")
kill_process("show-component-extension-options")
def update(self, sleep_time):
succeed = False
while True:
self.logger.debug("enter daily update process. %s" %
datetime.now().strftime('%Y-%m-%d %H:%M:%S'))
try:
if self.cal_client.is_trading_day():
#self.logger.info("is trading day. %s, succeed:%s" % (datetime.now().strftime('%Y-%m-%d %H:%M:%S'), succeed))
if self.is_collecting_time():
self.logger.debug(
"enter collecting time. %s, succeed:%s" %
(datetime.now().strftime('%Y-%m-%d %H:%M:%S'),
succeed))
if not succeed:
self.clear_network_env()
mdate = datetime.now().strftime('%Y-%m-%d')
ndate = get_latest_data_date()
if ndate is not None:
if ndate >= transfer_date_string_to_int(mdate):
if self.updating_date is None:
self.updating_date = mdate
succeed = self.bootstrap(
cdate=self.updating_date,
exec_date=self.updating_date)
if succeed: self.updating_date = None
else:
self.logger.debug("%s is older for %s" %
(ndate, mdate))
else:
succeed = False
gevent.sleep(sleep_time)
except Exception as e:
time.sleep(1)
self.logger.error(e)
def init_combination_info(self):
trading_info = self.comb_info_client.get()
for _, code_id in trading_info['code'].iteritems():
if str(code_id) not in self.combination_objs:
self.combination_objs[str(code_id)] = Combination(
code_id, self.dbinfo)
def init_base_float_profit(self):
def _set_base_float_profit(code_id):
if CStock(code_id).set_base_floating_profit():
self.logger.info("%s set base float profit success" % code_id)
return (code_id, True)
else:
self.logger.error("%s set base float profit failed" % code_id)
return (code_id, False)
df = self.stock_info_client.get()
if df.empty: return False
failed_list = df.code.tolist()
return process_concurrent_run(_set_base_float_profit,
failed_list,
num=8)
def init_rindex_valuation_info(self, cdate):
for code in ct.INDEX_DICT:
if not self.cvaluation_client.set_index_valuation(code, cdate):
self.logger.error(
"{} set {} data for rvaluation failed".format(code, mdate))
return False
return True
def init_rvaluation_info(self, cdate=None):
def cget(mdate, code):
return code, CStock(code).get_val_data(mdate)
df = self.stock_info_client.get()
code_list = df.code.tolist()
try:
obj_pool = Pool(5000)
all_df = pd.DataFrame()
cfunc = partial(cget, cdate)
for code_data in obj_pool.imap_unordered(cfunc, code_list):
if code_data[1] is not None and not code_data[1].empty:
tem_df = code_data[1]
tem_df['code'] = code_data[0]
all_df = all_df.append(tem_df)
obj_pool.join(timeout=5)
obj_pool.kill()
all_df = all_df.reset_index(drop=True)
file_name = "{}.csv".format(cdate)
file_path = Path(ct.RVALUATION_DIR) / file_name
all_df.to_csv(file_path,
index=False,
header=True,
mode='w',
encoding='utf8')
return True
except Exception as e:
self.logger.error(e)
return False
def init_valuation_info(self, cdate=None):
df = self.stock_info_client.get()
code_list = df['code'].tolist()
time2market_list = df['timeToMarket'].tolist()
code2timedict = dict(zip(code_list, time2market_list))
cfun = partial(self.cvaluation_client.set_stock_valuation,
code2timedict, cdate)
return process_concurrent_run(cfun,
code_list,
num=15,
black_list=list())
def init_stock_info(self, cdate=None):
def _set_stock_info(mdate, bonus_info, index_info, code_id):
try:
if CStock(code_id).set_k_data(bonus_info, index_info, mdate):
self.logger.info("%s set k data success for date:%s",
code_id, mdate)
return (code_id, True)
else:
self.logger.error("%s set k data failed for date:%s",
code_id, mdate)
return (code_id, False)
except Exception as e:
self.logger.error("%s set k data for date %s exception:%s",
code_id, mdate, e)
return (code_id, False)
#get stock bonus info
bonus_info = pd.read_csv("/data/tdx/base/bonus.csv",
sep=',',
dtype={
'code': str,
'market': int,
'type': int,
'money': float,
'price': float,
'count': float,
'rate': float,
'date': int
})
index_info = CIndex('000001').get_k_data()
if index_info is None or index_info.empty: return False
df = self.stock_info_client.get()
if df.empty: return False
failed_list = df.code.tolist()
if cdate is None:
cfunc = partial(_set_stock_info, cdate, bonus_info, index_info)
return process_concurrent_run(cfunc, failed_list, num=8)
else:
cfunc = partial(_set_stock_info, cdate, bonus_info, index_info)
succeed = True
if not process_concurrent_run(cfunc, failed_list, num=8):
succeed = False
return succeed
#start_date = get_day_nday_ago(cdate, num = 4, dformat = "%Y-%m-%d")
#for mdate in get_dates_array(start_date, cdate, asending = True):
# if self.cal_client.is_trading_day(mdate):
# self.logger.info("start recording stock info: %s", mdate)
# cfunc = partial(_set_stock_info, mdate, bonus_info, index_info)
# if not process_concurrent_run(cfunc, failed_list, num = 500):
# self.logger.error("compute stock info for %s failed", mdate)
# return False
#return True
def init_industry_info(self, cdate, num=1):
def _set_industry_info(cdate, code_id):
return (code_id, CIndex(code_id).set_k_data(cdate))
df = self.industry_info_client.get()
if cdate is None:
cfunc = partial(_set_industry_info, cdate)
return concurrent_run(cfunc, df.code.tolist(), num=5)
else:
succeed = True
start_date = get_day_nday_ago(cdate, num=num, dformat="%Y-%m-%d")
for mdate in get_dates_array(start_date, cdate, asending=True):
if self.cal_client.is_trading_day(mdate):
cfunc = partial(_set_industry_info, mdate)
if not concurrent_run(cfunc, df.code.tolist(), num=5):
succeed = False
return succeed
def init_yesterday_hk_info(self, cdate, num):
succeed = True
for data in ((ct.SH_MARKET_SYMBOL, ct.HK_MARKET_SYMBOL),
(ct.SZ_MARKET_SYMBOL, ct.HK_MARKET_SYMBOL)):
if not self.connect_client.set_market(data[0], data[1]):
self.logger.error("connect_client for %s failed" % data)
succeed = False
continue
if not self.connect_client.update(cdate, num=num):
succeed = False
self.connect_client.close()
self.connect_client.quit()
return succeed
def get_concerned_index_codes(self):
index_codes = list(ct.INDEX_DICT.keys())
#添加MSCI板块
index_codes.append('880883')
return index_codes
def init_index_components_info(self, cdate=None):
if cdate is None: cdate = datetime.now().strftime('%Y-%m-%d')
def _set_index_info(code_id):
if code_id in self.index_objs:
_obj = self.index_objs[code_id]
else:
_obj = CIndex(code_id) if code_id in list(
ct.INDEX_DICT.keys()) else TdxFgIndex(code_id)
return (code_id, _obj.set_components_data(cdate))
index_codes = self.get_concerned_index_codes()
return concurrent_run(_set_index_info, index_codes, num=10)
def set_bull_stock_ratio(self, cdate, num=10):
def _set_bull_stock_ratio(code_id):
return (code_id, BullStockRatio(code_id).update(cdate, num))
index_codes = self.get_concerned_index_codes()
return concurrent_run(_set_bull_stock_ratio, index_codes)
def init_tdx_index_info(self, cdate=None, num=1):
def _set_index_info(cdate, code_id):
try:
if code_id in self.index_objs:
_obj = self.index_objs[code_id]
else:
_obj = CIndex(code_id) if code_id in list(
ct.TDX_INDEX_DICT.keys()) else TdxFgIndex(code_id)
return (code_id, _obj.set_k_data(cdate))
except Exception as e:
self.logger.error(e)
return (code_id, False)
#index_code_list = self.get_concerned_index_codes()
index_code_list = list(ct.TDX_INDEX_DICT.keys())
if cdate is None:
cfunc = partial(_set_index_info, cdate)
return concurrent_run(cfunc, index_code_list, num=5)
else:
succeed = True
start_date = get_day_nday_ago(cdate, num=num, dformat="%Y-%m-%d")
for mdate in get_dates_array(start_date, cdate, asending=True):
if self.cal_client.is_trading_day(mdate):
cfunc = partial(_set_index_info, mdate)
if not concurrent_run(cfunc, index_code_list, num=5):
succeed = False
return succeed
class DataManager:
def __init__(self, dbinfo=ct.DB_INFO, redis_host=None):
self.dbinfo = dbinfo
self.logger = getLogger(__name__)
self.index_objs = dict()
self.stock_objs = dict()
self.combination_objs = dict()
self.cal_client = CCalendar(dbinfo, redis_host)
self.index_info_client = IndexInfo()
self.comb_info_client = CombinationInfo(dbinfo, redis_host)
self.stock_info_client = CStockInfo(dbinfo, redis_host)
self.rindex_stock_data_client = RIndexStock(dbinfo, redis_host)
self.industry_info_client = IndustryInfo(dbinfo, redis_host)
self.rindustry_info_client = RIndexIndustryInfo(dbinfo, redis_host)
self.limit_client = CLimit(dbinfo, redis_host)
self.animation_client = CAnimation(dbinfo, redis_host)
self.subscriber = Subscriber()
self.quote_handler = StockQuoteHandler()
self.ticker_handler = TickerHandler()
self.connect_client = StockConnect(market_from=ct.SH_MARKET_SYMBOL,
market_to=ct.HK_MARKET_SYMBOL,
dbinfo=dbinfo,
redis_host=redis_host)
self.margin_client = Margin(dbinfo=dbinfo, redis_host=redis_host)
self.emotion_client = Emotion(dbinfo=dbinfo, redis_host=redis_host)
self.sh_exchange_client = StockExchange(ct.SH_MARKET_SYMBOL)
self.sz_exchange_client = StockExchange(ct.SZ_MARKET_SYMBOL)
def is_collecting_time(self, now_time=datetime.now()):
_date = now_time.strftime('%Y-%m-%d')
y, m, d = time.strptime(_date, "%Y-%m-%d")[0:3]
aft_open_hour, aft_open_minute, aft_open_second = (19, 00, 00)
aft_open_time = datetime(y, m, d, aft_open_hour, aft_open_minute,
aft_open_second)
aft_close_hour, aft_close_minute, aft_close_second = (23, 59, 59)
aft_close_time = datetime(y, m, d, aft_close_hour, aft_close_minute,
aft_close_second)
return aft_open_time < now_time < aft_close_time
def is_morning_time(self, now_time=datetime.now()):
_date = now_time.strftime('%Y-%m-%d')
y, m, d = time.strptime(_date, "%Y-%m-%d")[0:3]
mor_open_hour, mor_open_minute, mor_open_second = (0, 0, 0)
mor_open_time = datetime(y, m, d, mor_open_hour, mor_open_minute,
mor_open_second)
mor_close_hour, mor_close_minute, mor_close_second = (6, 30, 0)
mor_close_time = datetime(y, m, d, mor_close_hour, mor_close_minute,
mor_close_second)
return mor_open_time < now_time < mor_close_time
def collect_combination_runtime_data(self):
def _combination_run(code_id):
self.combination_objs[code_id].run()
return (code_id, True)
todo_iplist = list(self.combination_objs.keys())
return concurrent_run(_combination_run, todo_iplist, num=10)
def collect_stock_runtime_data(self):
if self.ticker_handler.empty(): return
datas = self.ticker_handler.getQueue()
while not datas.empty():
df = datas.get()
df = df.set_index('time')
df.index = pd.to_datetime(df.index)
for code_str in set(df.code):
code_id = code_str.split('.')[1]
self.stock_objs[code_id].run(df.loc[df.code == code_str])
def init_real_stock_info(self):
concerned_list = self.comb_info_client.get_concerned_list()
prefix_concerned_list = [add_prifix(code) for code in concerned_list]
ret = self.subscriber.subscribe(prefix_concerned_list, SubType.TICKER,
self.ticker_handler)
if 0 == ret:
for code in concerned_list:
if code not in self.stock_objs:
self.stock_objs[code] = CStock(code,
self.dbinfo,
should_create_influxdb=True,
should_create_mysqldb=True)
return ret
def init_index_info(self):
index_list = ct.INDEX_DICT.keys()
prefix_index_list = [add_index_prefix(code) for code in index_list]
ret = self.subscriber.subscribe(prefix_index_list, SubType.QUOTE,
self.quote_handler)
if 0 != ret:
self.logger.error("subscribe for index list failed")
return ret
for code in index_list:
if code not in self.index_objs:
self.index_objs[code] = CIndex(code,
should_create_influxdb=True,
should_create_mysqldb=True)
def collect_index_runtime_data(self):
if self.quote_handler.empty(): return
datas = self.quote_handler.getQueue()
while not datas.empty():
df = datas.get()
df['time'] = df.data_date + ' ' + df.data_time
df = df.drop(['data_date', 'data_time'], axis=1)
df = df.set_index('time')
df.index = pd.to_datetime(df.index)
for code_str in set(df.code):
code_id = code_str.split('.')[1]
self.index_objs[code_id].run(df.loc[df.code == code_str])
def run(self, sleep_time):
while True:
try:
if self.cal_client.is_trading_day():
if is_trading_time():
sleep_time = 1
if not self.subscriber.status():
self.subscriber.start()
if 0 == self.init_index_info(
) and 0 == self.init_real_stock_info():
self.init_combination_info()
else:
self.logger.debug("enter stop dict time")
self.subscriber.stop()
else:
self.collect_stock_runtime_data()
self.collect_combination_runtime_data()
self.collect_index_runtime_data()
self.animation_client.collect()
else:
sleep_time = 60
if self.subscriber.status():
self.subscriber.stop()
except Exception as e:
traceback.print_exc()
self.logger.error(e)
time.sleep(sleep_time)
def set_update_info(self,
step_length,
exec_date,
cdate=None,
filename=ct.STEPFILE):
step_info = dict()
if cdate is None: cdate = 'none'
step_info[cdate] = dict()
step_info[cdate]['step'] = step_length
step_info[cdate]['date'] = exec_date
with open(filename, 'w') as f:
json.dump(step_info, f)
self.logger.info("finish step :%s" % step_length)
def get_update_info(self,
cdate=None,
exec_date=None,
filename=ct.STEPFILE):
if cdate is None: cdate = 'none'
if not os.path.exists(filename): return (0, exec_date)
with open(filename, 'r') as f:
step_info = json.load(f)
if cdate not in step_info: return (0, exec_date)
return (step_info[cdate]['step'], step_info[cdate]['date'])
def bootstrap(self,
cdate=None,
exec_date=datetime.now().strftime('%Y-%m-%d')):
finished_step, exec_date = self.get_update_info(cdate, exec_date)
self.logger.info("enter updating.%s" % finished_step)
if finished_step < 1:
if not self.cal_client.init():
self.logger.error("cal_client init failed")
return False
self.set_update_info(1, exec_date, cdate)
if finished_step < 2:
if not self.index_info_client.update():
self.logger.error("index_info init failed")
return False
self.set_update_info(2, exec_date, cdate)
if finished_step < 3:
if not self.stock_info_client.update():
self.logger.error("stock_info init failed")
return False
self.set_update_info(3, exec_date, cdate)
if finished_step < 4:
if not self.comb_info_client.update():
self.logger.error("comb_info init failed")
return False
self.set_update_info(4, exec_date, cdate)
if finished_step < 5:
if not self.industry_info_client.update():
self.logger.error("industry_info init failed")
return False
self.set_update_info(5, exec_date, cdate)
if finished_step < 6:
if not self.download_and_extract(exec_date):
self.logger.error("download_and_extract failed")
return False
self.set_update_info(6, exec_date, cdate)
if finished_step < 7:
if not self.init_tdx_index_info(cdate):
self.logger.error("init_tdx_index_info failed")
return False
self.set_update_info(7, exec_date, cdate)
if finished_step < 8:
if not self.sh_exchange_client.update(exec_date, num=30):
self.logger.error("sh exchange update failed")
return False
self.set_update_info(8, exec_date, cdate)
if finished_step < 9:
if not self.sz_exchange_client.update(exec_date, num=30):
self.logger.error("sz exchange update failed")
return False
self.set_update_info(9, exec_date, cdate)
if finished_step < 10:
if not self.init_index_components_info(exec_date):
self.logger.error("init index components info failed")
return False
self.set_update_info(10, exec_date, cdate)
if finished_step < 11:
if not self.init_industry_info(cdate):
self.logger.error("init_industry_info failed")
return False
self.set_update_info(11, exec_date, cdate)
if finished_step < 12:
if not self.rindustry_info_client.update(exec_date):
self.logger.error("init %s rindustry info failed" % exec_date)
return False
self.set_update_info(12, exec_date, cdate)
if finished_step < 13:
if not self.limit_client.update(exec_date):
self.logger.error("init_limit_info failed")
return False
self.set_update_info(13, exec_date, cdate)
if finished_step < 14:
if not self.init_yesterday_hk_info(exec_date):
self.logger.error("init_yesterday_hk_info failed")
return False
self.set_update_info(14, exec_date, cdate)
if finished_step < 15:
if not self.margin_client.update(exec_date):
self.logger.error("init_yesterday_margin failed")
return False
self.set_update_info(15, exec_date, cdate)
if finished_step < 16:
if not self.init_stock_info(cdate):
self.logger.error("init_stock_info set failed")
return False
self.set_update_info(16, exec_date, cdate)
if finished_step < 17:
if not self.init_base_float_profit():
self.logger.error("init base float profit for all stock")
return False
self.set_update_info(17, exec_date, cdate)
if finished_step < 18:
if not self.rindex_stock_data_client.update(exec_date, num=300):
self.logger.error("rindex_stock_data set failed")
return False
self.set_update_info(18, exec_date, cdate)
self.logger.info("updating succeed")
return True
def update(self, sleep_time):
while True:
self.logger.info("enter daily update process. %s" %
datetime.now().strftime('%Y-%m-%d %H:%M:%S'))
try:
if self.cal_client.is_trading_day():
self.logger.info(
"is trading day. %s" %
datetime.now().strftime('%Y-%m-%d %H:%M:%S'))
if self.is_collecting_time():
self.logger.info(
"is collecting time. %s" %
datetime.now().strftime('%Y-%m-%d %H:%M:%S'))
self.bootstrap(
cdate=datetime.now().strftime('%Y-%m-%d'))
except Exception as e:
kill_process("google-chrome")
kill_process("renderer")
kill_process("Xvfb")
kill_process("zygote")
kill_process("defunct")
kill_process("show-component-extension-options")
self.logger.error(e)
time.sleep(sleep_time)
def init_combination_info(self):
trading_info = self.comb_info_client.get()
for _, code_id in trading_info['code'].iteritems():
if str(code_id) not in self.combination_objs:
self.combination_objs[str(code_id)] = Combination(
code_id, self.dbinfo)
def init_base_float_profit(self):
def _set_base_float_profit(code_id):
return (code_id,
True) if CStock(code_id).set_base_floating_profit() else (
code_id, False)
failed_list = self.stock_info_client.get().code.tolist()
return process_concurrent_run(_set_base_float_profit,
failed_list,
num=500)
def init_stock_info(self, cdate=None):
def _set_stock_info(_date, bonus_info, index_info, code_id):
try:
if CStock(code_id).set_k_data(bonus_info, index_info, _date):
self.logger.info("%s set k data success" % code_id)
return (code_id, True)
else:
self.logger.error("%s set k data failed" % code_id)
return (code_id, False)
except Exception as e:
self.logger.error("%s set k data exception:%s" % (code_id, e))
return (code_id, False)
#get stock bonus info
bonus_info = pd.read_csv("/data/tdx/base/bonus.csv",
sep=',',
dtype={
'code': str,
'market': int,
'type': int,
'money': float,
'price': float,
'count': float,
'rate': float,
'date': int
})
index_info = CIndex('000001').get_k_data()
if index_info is None or index_info.empty: return False
df = self.stock_info_client.get()
failed_list = df.code.tolist()
if cdate is None:
cfunc = partial(_set_stock_info, cdate, bonus_info, index_info)
return process_concurrent_run(cfunc, failed_list, num=5)
else:
succeed = True
start_date = get_day_nday_ago(cdate, num=10, dformat="%Y-%m-%d")
for mdate in get_dates_array(start_date, cdate, asending=True):
if self.cal_client.is_trading_day(mdate):
cfunc = partial(_set_stock_info, mdate, bonus_info,
index_info)
if not process_concurrent_run(cfunc, failed_list, num=500):
succeed = False
return succeed
def init_industry_info(self, cdate):
def _set_industry_info(cdate, code_id):
return (code_id, CIndex(code_id).set_k_data(cdate))
df = self.industry_info_client.get()
if cdate is None:
cfunc = partial(_set_industry_info, cdate)
return concurrent_run(cfunc, df.code.tolist(), num=5)
else:
succeed = True
start_date = get_day_nday_ago(cdate, num=30, dformat="%Y-%m-%d")
for mdate in get_dates_array(start_date, cdate, asending=True):
if self.cal_client.is_trading_day(mdate):
cfunc = partial(_set_industry_info, mdate)
if not concurrent_run(cfunc, df.code.tolist(), num=5):
succeed = False
return succeed
def init_yesterday_hk_info(self, cdate):
succeed = True
for data in ((ct.SH_MARKET_SYMBOL, ct.HK_MARKET_SYMBOL),
(ct.SZ_MARKET_SYMBOL, ct.HK_MARKET_SYMBOL)):
if not self.connect_client.set_market(data[0], data[1]):
self.logger.error("connect_client for %s failed" % data)
succeed = False
continue
if not self.connect_client.update(cdate):
succeed = False
self.connect_client.close()
self.connect_client.quit()
kill_process("zygote")
kill_process("defunct")
kill_process("show-component-extension-options")
return succeed
def init_index_components_info(self, cdate=None):
if cdate is None: cdate = datetime.now().strftime('%Y-%m-%d')
def _set_index_info(code_id):
_obj = self.index_objs[
code_id] if code_id in self.index_objs else CIndex(code_id)
return (code_id, _obj.set_components_data(cdate))
return concurrent_run(_set_index_info,
list(ct.INDEX_DICT.keys()),
num=10)
def init_tdx_index_info(self, cdate=None):
def _set_index_info(cdate, code_id):
try:
_obj = self.index_objs[
code_id] if code_id in self.index_objs else CIndex(code_id)
return (code_id, _obj.set_k_data(cdate))
except Exception as e:
self.logger.error(e)
return (code_id, False)
if cdate is None:
cfunc = partial(_set_index_info, cdate)
return concurrent_run(cfunc, list(ct.TDX_INDEX_DICT.keys()), num=5)
else:
succeed = True
start_date = get_day_nday_ago(cdate, num=30, dformat="%Y-%m-%d")
for mdate in get_dates_array(start_date, cdate, asending=True):
if self.cal_client.is_trading_day(mdate):
cfunc = partial(_set_index_info, mdate)
if not concurrent_run(
cfunc, list(ct.TDX_INDEX_DICT.keys()), num=5):
succeed = False
return succeed
def download_and_extract(self, cdate):
try:
if not download(ct.ZIP_DIR, cdate): return False
list_files = os.listdir(ct.ZIP_DIR)
for filename in list_files:
if not filename.startswith('.'):
file_path = os.path.join(ct.ZIP_DIR, filename)
if os.path.exists(file_path):
unzip(file_path, ct.TIC_DIR)
return True
except Exception as e:
self.logger.error(e)
return False
def relation_plot(self, df, good_list):
close_price_list = [
df[df.code == code].close.tolist() for code in good_list
]
close_prices = np.vstack(close_price_list)
open_price_list = [
df[df.code == code].open.tolist() for code in good_list
]
open_prices = np.vstack(open_price_list)
# the daily variations of the quotes are what carry most information
variation = (close_prices - open_prices) * 100 / open_prices
logger.info("get variation succeed")
# #############################################################################
# learn a graphical structure from the correlations
edge_model = covariance.GraphLassoCV()
# standardize the time series: using correlations rather than covariance is more efficient for structure recovery
X = variation.copy().T
X /= X.std(axis=0)
edge_model.fit(X)
logger.info("mode compute succeed")
# #############################################################################
# cluster using affinity propagation
_, labels = cluster.affinity_propagation(edge_model.covariance_)
n_labels = labels.max()
code_list = np.array(good_list)
industry_dict = dict()
industry_df_info = IndustryInfo.get()
for index, name in industry_df_info.name.iteritems():
content = industry_df_info.loc[index]['content']
a_code_list = json.loads(content)
for code in a_code_list:
industry_dict[code] = name
cluster_dict = dict()
for i in range(n_labels + 1):
cluster_dict[i] = code_list[labels == i]
name_list = [
CStockInfo.get(code, 'name') for code in code_list[labels == i]
]
logger.info('cluster code %i: %s' %
((i + 1), ', '.join(name_list)))
cluster_info = dict()
for group, _code_list in cluster_dict.items():
for code in _code_list:
iname = industry_dict[code]
if group not in cluster_info: cluster_info[group] = set()
cluster_info[group].add(iname)
logger.info('cluster inustry %i: %s' %
((i + 1), ', '.join(list(cluster_info[group]))))
# #############################################################################
# find a low-dimension embedding for visualization: find the best position of
# the nodes (the stocks) on a 2D plane
# we use a dense eigen_solver to achieve reproducibility (arpack is
# initiated with random vectors that we don't control). In addition, we
# use a large number of neighbors to capture the large-scale structure.
node_position_model = manifold.LocallyLinearEmbedding(
n_components=2, eigen_solver='dense', n_neighbors=6)
embedding = node_position_model.fit_transform(X.T).T
# #############################################################################
# visualizatio
plt.figure(1, facecolor='w', figsize=(10, 8))
plt.clf()
ax = plt.axes([0., 0., 1., 1.])
plt.axis('off')
# display a graph of the partial correlations
partial_correlations = edge_model.precision_.copy()
d = 1 / np.sqrt(np.diag(partial_correlations))
partial_correlations *= d
partial_correlations *= d[:, np.newaxis]
non_zero = (np.abs(np.triu(partial_correlations, k=1)) > 0.02)
# plot the nodes using the coordinates of our embedding
plt.scatter(embedding[0],
embedding[1],
s=100 * d**2,
c=labels,
cmap=plt.cm.nipy_spectral)
# plot the edges
start_idx, end_idx = np.where(non_zero)
# a sequence of (*line0*, *line1*, *line2*), where:: linen = (x0, y0), (x1, y1), ... (xm, ym)
segments = [[embedding[:, start], embedding[:, stop]]
for start, stop in zip(start_idx, end_idx)]
values = np.abs(partial_correlations[non_zero])
lc = LineCollection(segments,
zorder=0,
cmap=plt.cm.hot_r,
norm=plt.Normalize(0, .7 * values.max()))
lc.set_array(values)
lc.set_linewidths(15 * values)
ax.add_collection(lc)
# add a label to each node. The challenge here is that we want to position the labels to avoid overlap with other labels
for index, (name, label,
(x, y)) in enumerate(zip(code_list, labels, embedding.T)):
dx = x - embedding[0]
dx[index] = 1
dy = y - embedding[1]
dy[index] = 1
this_dx = dx[np.argmin(np.abs(dy))]
this_dy = dy[np.argmin(np.abs(dx))]
if this_dx > 0:
horizontalalignment = 'left'
x = x + .002
else:
horizontalalignment = 'right'
x = x - .002
if this_dy > 0:
verticalalignment = 'bottom'
y = y + .002
else:
verticalalignment = 'top'
y = y - .002
plt.text(x,
y,
name,
size=10,
horizontalalignment=horizontalalignment,
verticalalignment=verticalalignment,
bbox=dict(facecolor='w',
edgecolor=plt.cm.nipy_spectral(label /
float(n_labels)),
alpha=.6))
plt.xlim(
embedding[0].min() - .15 * embedding[0].ptp(),
embedding[0].max() + .10 * embedding[0].ptp(),
)
plt.ylim(embedding[1].min() - .03 * embedding[1].ptp(),
embedding[1].max() + .03 * embedding[1].ptp())
plt.savefig('/tmp/relation.png', dpi=1000)
def get_industry(self, code):
"""获取沪深股股票通达信行业信息"""
rdf = IndustryInfo.get_industry()
rdf = rdf[rdf.content.str.contains(code)]
return rdf['name'].values[0] if not rdf.empty else None
import sys
from os.path import abspath, dirname
sys.path.insert(0, dirname(dirname(dirname(abspath(__file__)))))
import traceback
import const as ct
import pandas as pd
from rstock import RIndexStock
from cstock_info import CStockInfo
from industry_info import IndustryInfo
if __name__ == '__main__':
try:
mdate = '2019-08-02'
cobj = CStockInfo()
robj = RIndexStock()
iobj = IndustryInfo()
black_list = list(ct.BLACK_DICT.keys())
bdf = cobj.get()
stock_info = robj.get_data(mdate)
idf = iobj.get_csi_industry_data(mdate)
df = pd.merge(bdf, idf, how='left', on=['code'])
df = pd.merge(stock_info, df, how='inner', on=['code'])
df = df[~df.code.isin(black_list)]
df = df[(df.profit > 1) & (df.profit < 3) & (df.pday > 30) &
(df.timeToMarket < 20150101)]
df = df.reset_index(drop=True)
#df = df[['code', 'name', 'industry', 'profit', 'pday', 'pind_name', 'sind_name', 'tind_name', 'find_name']]
df = df[['code', 'name', 'profit', 'pday', 'find_name']]
for name, contains in df.groupby('find_name'):
if len(contains) > 2: