def updatePotInfo():
"""
:param df_H_L_Pot:
:return:
"""
# h_l_pot_info_url = '.\InfoRestore\df_H_L_Pot.pkl'
if os.path.exists(h_l_pot_info_url):
with open(h_l_pot_info_url, 'rb') as f:
try:
h_l_pot_info = pickle.load(f)
except:
h_l_pot_info = pd.DataFrame()
else:
h_l_pot_info = pd.DataFrame()
if h_l_pot_info.empty:
h_l_pot_info = initPotInfo(get_h_l_pot(stk_list))
else:
new_pot_info = get_h_l_pot(stk_list)
h_l_pot_info = pd.concat([
h_l_pot_info.drop([
'half_year_high', 'half_year_low', 'month_high', 'month_low',
'year_high', 'year_low'
], 1),
new_pot_info.drop('stk', 1)
],
axis=1)
# 将结果序列化
with open(h_l_pot_info_url, 'wb') as f:
pickle.dump(h_l_pot_info, f)
send_qq(u'影子', '高底线信息更新成功！')
def judgeAndSendMsg():
"""
按频率调用，
:return:
"""
if os.path.exists(h_l_pot_info_url):
with open(h_l_pot_info_url, 'rb') as f:
h_l_pot_info = pickle.load(f)
else:
print('函数 judgeAndSendMsg: 加载高低信息失败！')
return
df_H_L_Pot = h_l_pot_info
for stk in df_H_L_Pot.index:
# 获取该股票的实时价格
current_price = float(
ts.get_realtime_quotes(df_H_L_Pot.loc[stk,
'stk'])['price'].values[0])
# 将当前价格保存，用于验证计算准确性
df_H_L_Pot.loc[stk, 'current_price'] = current_price
""" 年线判断 """
df_H_L_Pot = lineJudge(df_H_L_Pot_index=stk,
current_price=current_price,
df_info=df_H_L_Pot,
line_str='year')
""" ----------------- 半年线判断 -----------------"""
df_H_L_Pot = lineJudge(df_H_L_Pot_index=stk,
current_price=current_price,
df_info=df_H_L_Pot,
line_str='half_year')
""" ----------------- 月线判断 ----------------"""
df_H_L_Pot = lineJudge(df_H_L_Pot_index=stk,
current_price=current_price,
df_info=df_H_L_Pot,
line_str='month')
""" 检查并发送消息 """
for idx in df_H_L_Pot.index:
# 遍历年线、半年线和月线
for sts in ['year_status', 'half_year_status', 'month_status']:
if (df_H_L_Pot.loc[idx, sts] != u'正常') & (df_H_L_Pot.loc[
idx, sts] != df_H_L_Pot.loc[idx, sts + '_last']):
send_qq(
u'影子', 'stk:' + getNameByStkCode(
g_total_stk_info_mysql, df_H_L_Pot.loc[idx, 'stk']) +
'\n' + '当前价格：' +
str(df_H_L_Pot.loc[idx, 'current_price']) + '\n' +
'事件： “' + df_H_L_Pot.loc[idx, sts + '_last'] + '” --> “' +
df_H_L_Pot.loc[idx, sts] + '”' + '\n\n')
df_H_L_Pot.loc[idx, sts + '_last'] = df_H_L_Pot.loc[idx, sts]
with open(h_l_pot_info_url, 'wb') as f:
pickle.dump(h_l_pot_info, f)
print('函数 judgeAndSendMsg: 完成本次判断！')
def update_price_ratio_info():
"""
:param df_H_L_Pot:
:return:
"""
# 将结果序列化
with open(price_ratio_info_url, 'wb') as f:
pickle.dump(h_l_pot_info, f)
send_qq(u'影子', '高底线信息更新成功！')
def updateConcernStkMData():
"""
定时器定时调用，更新各stk的M离心度历史数据
:return:
"""
try:
for stk in stk_list:
saveStkMRankHistoryData(stk_code=stk,
history_days=400,
m_days=9,
save_dir='./M_data/')
send_qq('影子', '更新' + stk + '离心度历史数据成功！')
except:
send_qq('影子', '更新离心度历史数据失败！')
def checkDivergeLowLevel():
"""
供定时器调用的回调函数，按频率检查关心的stk的，对高于80分的进行提示
:return:
"""
for stk in stk_list:
r = calRealtimeRank(stk_code=stk,
M_days=9,
history_data_dir=MDataPWD + '/M_data/')
if r > 80:
send_qq('影子',
'Attention：\n' + stk + '趋向高分！分数为：' + str('%0.2f') % r)
else:
print(stk + '分数处于正常状态！分数为：' + str('%0.2f') % r)
def JudgeSingleStk(stk_code, stk_price_last, stk_amount_last, earn_threshold):
# 获取该股票的实时价格
current_price = float(ts.get_realtime_quotes(stk_code)['price'].values[0])
price_diff = current_price - stk_price_last
if current_price == 0.0:
print(stk_code + 'price==0.0! 返回！')
return
buy_amount = math.floor((money_each_opt / current_price) / 100) * 100
# 计算其离心度分数
try:
rank9 = calRealtimeRank(stk_code=stk_code,
M_days=9,
history_data_dir=MDataPWD + '/M_data/')
except:
rank9 = -1
if (price_diff * stk_amount_last > earn_threshold * remind_ratio) & (
price_diff * stk_amount_last < earn_threshold):
send_qq(
'影子', "Near! S! " + stk_code + '\nAmount:' + str(stk_amount_last) +
'\nP_now:' + str(current_price) + '\nP_last:' +
str(stk_price_last) + '\nthreshold:' + str(earn_threshold) +
'\nM9_rank:' + str('%0.2f' % rank9))
elif price_diff * stk_amount_last > earn_threshold:
send_qq(
'影子', "Reach! S! " + stk_code + '\nAmount:' +
str(stk_amount_last) + '\nP_now:' + str(current_price) +
'\nP_last:' + str(stk_price_last) + '\nthreshold:' +
str(earn_threshold) + '\nM9_rank:' + str('%0.2f' % rank9))
elif (price_diff * buy_amount > -earn_threshold) & (
price_diff * buy_amount < -earn_threshold * remind_ratio):
send_qq(
'影子', "Near! B! " + stk_code + '\nAmount:' + str(buy_amount) +
'\nP_now:' + str(current_price) + '\nP_last:' +
str(stk_price_last) + '\nthreshold:' + str(earn_threshold) +
'\nM9_rank:' + str('%0.2f' % rank9))
elif price_diff * buy_amount < -earn_threshold:
send_qq(
'影子', "Reach! B! " + stk_code + '\nAmount:' + str(buy_amount) +
'\nP_now:' + str(current_price) + '\nP_last:' +
str(stk_price_last) + '\nthreshold:' + str(earn_threshold) +
'\nM9_rank:' + str('%0.2f' % rank9))
else:
print('未触发任何警戒线！')
def checkDivergeLowLevel_Sea():
"""
供定时器调用的回调函数，按频率检查关心的stk的，对高于80分的进行提示
:return:
"""
for stk in stk_pool:
try:
r = calRealtimeRank(stk_code=stk,
M_days=9,
history_data_dir=SeaSelectDataPWD +
'/stk_pool_data/')
if r > 94:
send_qq('影子',
'Attention：\n' + stk + '趋向高分！分数为：' + str('%0.2f') % r)
else:
print(stk + '分数处于正常状态！分数为：' + str('%0.2f') % r)
except:
print(stk + '判断时出现异常！')
def saveWeightFile():
"""
将持仓股票的权值存为文件
:param dir:
:return:
"""
# 新方式：lacalDBInfo包括除了“数据库名”之外的其他参数
localDBInfo = {
'host': 'localhost',
'port': 3306,
'user': '******',
'password': '******',
'charset': 'utf8'
}
db_name = 'stk_opt_info'
table_history = 'history'
table_now = 'now'
(conn_opt, engine_opt) = genDbConn(localDBInfo, db_name)
df = pd.read_sql(con=conn_opt, sql='select * from now')
if not df.empty:
df['weight'] = df.apply(lambda x: calWeight(x['stk_code']), axis=1)
dumpPickle(data=df, saveLocation='./Weight/', fileName='weight')
send_qq('影子',
'更新股票的权值成功！\n' + str(df.loc[:, ['stk_name', 'weight']].values))
else:
dumpPickle(data=pd.DataFrame(),
saveLocation='./Weight/',
fileName='weight')
send_qq('影子', '更新股票权值时发现 now 表为空！')
conn_opt.close()
