class FuturesSpider:
def __init__(self):
self.dm = DBManager("Futures_d_table")
def init_table(self):
for item in symbol_list:
self.dm.add_one({
"code": item.split("/")[0],
"symbol": item.split("/")[1],
"details": []
})
def start_crawl(self):
for symbol in symbol_list:
url = "http://stock2.finance.sina.com.cn/futures/api/json.php/IndexService.getInnerFuturesDailyKLine?symbol=" + symbol.split(
"/")[0]
print(url)
max_try = 8
for tries in range(max_try):
try:
content = requests.get(url)
self.parse_pager(content.content, symbol.split("/")[0])
break
except Exception:
if tries < (max_try - 1):
sleep(2)
continue
else:
print(symbol, "fail")
def parse_pager(self, content, code):
timer_list = [
x["date"]
for x in self.dm.find_one_by_key({"code": code})["details"]
]
data = json.loads(content)
for item in data:
__dict = {
"date": item[0],
"open": item[1],
"high": item[2],
"low": item[3],
"close": item[4],
"count": item[5]
}
if __dict["date"] not in timer_list:
self.dm.add_futures_item(code, __dict)
print(code, "success")
class EmaManager:
def __init__(self):
self.db_manager_tk = DBManager("fcr_w_details")
def get_buy_list(self, date):
code_list = [x for x in self.db_manager_tk.get_code_list_02()]
buy_list = list()
for code_item in code_list:
ticker = code_item["ticker"]
# 获取数据
close_list = list()
tk_details = self.db_manager_tk.find_one_by_key({"ticker": ticker})
for tk_item in [x for x in tk_details["price_list"]]:
if time_cmp(str(date), tk_item["date"]):
close_list.append(float(tk_item["close"]))
# 执行判断条件
if len(close_list) > 20:
ema_20_1 = np.mean(close_list[-20:])
ema_20_2 = np.mean(close_list[-21:-1])
if close_list[-1] > ema_20_1 and close_list[-2] < ema_20_2:
buy_list.append(ticker)
return buy_list
def get_sell_list(self, date):
code_list = [x for x in self.db_manager_tk.get_code_list_02()]
sell_list = list()
for code_item in code_list:
ticker = code_item["ticker"]
# 获取数据
close_list = list()
tk_details = self.db_manager_tk.find_one_by_key({"ticker": ticker})
for tk_item in [x for x in tk_details["price_list"]]:
if time_cmp(str(date), tk_item["date"]):
close_list.append(float(tk_item["close"]))
# 执行判断条件
if len(close_list) > 20:
ema_10_1 = np.mean(close_list[-10:])
ema_10_2 = np.mean(close_list[-11:-1])
if close_list[-1] < ema_10_1 and close_list[-2] > ema_10_2:
sell_list.append(ticker)
return sell_list
def fun_01(self):
code_list = [x for x in self.db_manager_tk.get_code_list_02()]
result_list = list()
for code_item in code_list:
ticker = code_item["ticker"]
# 获取数据
tk_details = self.db_manager_tk.find_one_by_key({"ticker": ticker})
close_list = list()
open_list = list()
high_list = list()
low_list = list()
volume_list = list()
for x in tk_details["price_list"]:
if x["close"] != "":
close_list.append(float(x["close"]))
open_list.append(float(x["open"]))
high_list.append(float(x["high"]))
low_list.append(float(x["low"]))
volume_list.append(float(x["volume"]))
ema_list = list()
for i in range(len(close_list)):
if i >= 20:
ema_list.append(np.mean(close_list[i - 20:i]))
else:
ema_list.append(close_list[i])
# 执行判断条件
for date in range(len(close_list) - 1):
# 均线呈现向上趋势
if ema_list[date] > ema_list[date - 4]:
# K线上穿且连续8周位于均线之下
if close_list[date] > ema_list[date]:
if close_list[date - 1] < ema_list[
date - 1] and close_list[date - 2] < ema_list[
date -
2] and close_list[date - 3] < ema_list[
date - 3] and close_list[
date - 4] < ema_list[date - 4]:
profit_rate = (
close_list[date + 1] -
open_list[date + 1]) / open_list[date + 1]
result_list.append(profit_rate)
print(len(result_list))
result_list_up = [x for x in result_list if x > 0]
result_list_down = [x for x in result_list if x < 0]
print(len(result_list_up), max(result_list_up),
np.mean(result_list_up))
print(len(result_list_down), min(result_list_down),
np.mean(result_list_down))
# 绘图
plt.subplot(111)
lable_x = np.arange(len(result_list))
lable_y = [x * 0 for x in range(len(result_list))]
# 绘制中轴线
plt.plot(lable_x,
lable_y,
color="#404040",
linewidth=1.0,
linestyle="-")
plt.bar(lable_x, result_list, color="g", width=1.0)
plt.xlim(lable_x.min(), lable_x.max() * 1.1)
plt.ylim(min(result_list) * 1.1, max(result_list) * 1.1)
plt.grid(True)
plt.show()
def fun_get_strategy(self):
pass