-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathdataMonthRetriever.py
215 lines (197 loc) · 11.1 KB
/
dataMonthRetriever.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
# pylint: disable=I0011,C0103,C0326,C0301, W0401,W0614
from cassandra.cluster import Cluster
from WindPy import *
import time
import datetime
# import logWriter
def monthRetrieve(startTime, endTime = datetime.datetime.today(),
fields1 = ['trade_status','close', 'mkt_freeshares','mkt_cap_float', 'mfd_buyamt_d', 'mfd_sellamt_d', 'roa', 'pe', 'pb'],
option1 = "ruleType=8;unit=1;traderType=1;Period=M;Fill=Previous;PriceAdj=B", multi_mfd = True):
# cassandra connect
cluster = Cluster(['192.168.1.111'])
session = cluster.connect('factors') # factors: factors_month
# 启动Wind API
w.start()
# 获取可交易日
times = w.tdays(startTime, endTime, "Period=M").Times
timeList = []
for i in range(len(times)):
row = str(times[i])
row = row[:row.find(' ')]
timeList.append(row)
print(timeList)
# # 【解耦：迁移至stock.py，定期更新】判断数据有效性
# 获取某个月份所有可交易的A股 （如此的话每次一支股票只拿一个数据，分多个时间点去拿，请求数目过多，改成批量拉取一支股票
# 所有因子
# stocks = w.wset("SectorConstituent", u"sector=全部A股;field=wind_code")
# validStocks ={}
# # Total stock: 3183 [2017-04-13]
# print (time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()), "Total A stocks number: ", len(stocks.Data[0]))
# # stock status update statement
# updateStmt = session.prepare('''INSERT INTO stock_info(stock, ipo_date, trade_status) VALUES (?,?,?)''')
#
# #for stock in ["000852.SZ","603788.SH","603987.SH","603988.SH","603989.SH","603990.SH","603991.SH","603993.SH"]:
# #for stock in ["000852.SZ","603788.SH","603990.SH","603991.SH","603993.SH"]:
# for stock in stocks.Data[0]:
# ipo_status = w.wsd(stock, "ipo_date, trade_status", datetime.datetime.today())
# #print (ipo_status)
# try:
# days = (datetime.datetime.today() - ipo_status.Data[0][0]).days
# # trade_status 不能用一个变量表示，而是一个时序的因子，这里的0/1只能用区分IPO是否符合要求
# if days > 90 and ipo_status.Data[1][0] == "交易":
# # if days > 90:
# validStocks[stock] = ipo_status.Data[1][0]
# session.execute(updateStmt, (stock, ipo_status.Data[0][0], '1'))
# else:
# # set status 0
# session.execute(updateStmt, (stock, ipo_status.Data[0][0], '0'))
# print (" Set invalid data: ", stock, str(ipo_status.Data[0][0]))
# except TypeError:
# print (" -- Log TypeError at Stock: ", stock, " :\t", str(ipo_status.Data[0][0]))
# Valid: 2819 [2017-04-13]
# tradable stocks' collection
rows = session.execute('''SELECT stock, ipo_date FROM stock_info WHERE trade_status = '1' ALLOW FILTERING ''')
validStocks = {}
validStockCode = []
for row in rows:
#validStocks[row.stock] = row.ipo_date
validStockCode.append(row.stock)
validN = len(validStockCode)
print (time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) , " valid stocks' number: ", validN)
#print (validStocks)
## 拉取因子，分阶段拉取，拉完异步存DB
if multi_mfd == True:
# columns = fields1 + ['mfd_buyamt_d2', 'mfd_sellamt_d2','mfd_buyamt_d4', 'mfd_sellamt_d4']
columns = fields1 + ['mfd_buyamt_d1', 'mfd_sellamt_d1','mfd_buyamt_d4', 'mfd_sellamt_d4']
else:
columns = fields1
dataList = [] #创建数组
cnt = 0 #2401 #当前拉取了多少支股票
index = 0 #2401 #上一次dump的位置，主要目的是通过此索引找到该股票代码
CHUNK_SIZE = 300 #每一次异步dump的股票个数
print("---- Retrieving factors: ",columns)
preparedStmt = session.prepare('''INSERT INTO factors_month(stock, factor, time, value) VALUES (?,?,?,?)''')
print (time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) , " ------ Starting retrieving...")
# 拉取交易状态便于之后数据过滤
hasTradeStatus = False
if len(fields1) > 1 and fields1[0] == 'trade_status':
hasTradeStatus = True
## 遍历所有股票
pos = index
while pos < validN:
stock = validStockCode[pos]
# 为了获取剩下不成功的股票从index开始
# for m in range(index, len(validStockCode)):
# stock = validStockCode[m]
# 只取 IPO 之后的数据【需求变更，IPO之前的ROA也可能是有用的数据】
# start = startTime if startTime > ipo_date.date() else ipo_date.date()
start = startTime
# 同一个变量，参数不一样，需要分成几次拉取
wsd = w.wsd(stock, fields1, start, endTime, option1)
if wsd.ErrorCode != 0:
print("--------------------- ERROR IN WIND ------------\r\n ErrorCode：", wsd.ErrorCode, " Stock: ",stock)
wsd_data = wsd.Data
if multi_mfd == True:
fields2 = ['mfd_buyamt_d', 'mfd_sellamt_d']
# 机构流入/流出额
option2 = "unit=1;traderType=1;Period=M;Fill=Previous;PriceAdj=B"
# 大户流入/流出额
# option2 = "unit=1;traderType=2;Period=M;Fill=Previous;PriceAdj=B"
wsd_data = wsd_data + w.wsd(stock, fields2, start, endTime, option2).Data
# 散户流入/流出额
option3 = "unit=1;traderType=4;Period=M;Fill=Previous;PriceAdj=B"
wsd_data = wsd_data + w.wsd(stock, fields2, start, endTime, option3).Data
##【修改：计算动量模块单独移出来，为可扩展性】mmt = close_1 / close_2; 没有数据增长率为0
# mmt = []
# mmt.append(1)
# for i in range(1, len(wsd_data[0])):
# if wsd_data[0][i] is not None and wsd_data[0][i] != 0:
# mmt.append(wsd_data[0][i] / wsd_data[0][i-1])
# else:
# mmt.append(float('nan'))
# wsd_data.append(mmt)
dataList.append(wsd_data)
cnt += 1
pos += 1
#阶段性异步导出 dump data asynchronously, 300 stocks / round
if cnt % CHUNK_SIZE == 0:
for s in range(index, cnt):
# try to catch Exception: 'CWSDService: corrupted response.'
try:
for i in range(len(columns)):
for j in range(len(dataList[s - index][i])):
#print (validStocks[s],columns[i],timeList[j],dataList[s - index][i][j])
try:
value = dataList[s - index][i][j]
if hasTradeStatus == True and i == 0:
# 交易 状态作为一个因子
if value is not None and value == "交易":
value = 1
else:
value = 0
elif value is not None:
value = float(value)
else:
value = float('nan')
except (ValueError, TypeError, KeyError) as e:
value = float('nan')
print ("--Log ValueError in ", validStockCode[s],"\t",columns[i],"\t",str(timeList[j]),"\t",str(value))
print ("EXCEPTION: ",e)
print ("--------------------------------------------------------------------------")
except IndexError as e:
print ("--------------------------------------------------------------------------")
print("len s: %d, len i: %d, len j: %d ~ " %(cnt, len(columns),len(timeList)), (s-index,i,j))
print(e)
session.execute_async(preparedStmt, (validStockCode[s],columns[i],timeList[j], value))
except IndexError as e:
print ("--------------------------------------------------------------------------")
print("WIND RESPONSE CORRUPT, START OVER!!! ", e)
cnt = index
pos = index
break
#记录上一次导出数据位置，清空buffer
index = cnt
dataList = []
print(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) ,'------ Dump NO.%d end at stock %s \n' % (cnt, stock))
print ("---- Last chunk size: ", len(dataList))
print(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) ,'---------------- Pulling finished!\n')
# 最后的剩余数据插入cassandra
for s in range(index, cnt):
for i in range(len(columns)):
for j in range(len(dataList[s - index][i])):
#print (validStocks[s],columns[i],timeList[j],dataList[s - index][i][j])
try:
value = dataList[s - index][i][j]
if hasTradeStatus == True and i == 0:
if value is not None and value == "交易":
value = 1
else:
value = 0
elif value is not None:
value = float(value)
else:
value = float('nan')
except (ValueError, TypeError, KeyError) as e:
value = float('nan')
print ("--Log ValueError in ", validStockCode[s],"\t",columns[i],"\t",str(timeList[j]),"\t",str(value))
print ("EXCEPTION: ",e)
print ("--------------------------------------------------------------------------")
except IndexError as e:
print ("--------------------------------------------------------------------------")
print("len s: %d, len i: %d, len j: %d ~ " %(cnt, len(columns),len(timeList)), (s-index,i,j))
print(e)
session.execute_async(preparedStmt, (validStockCode[s],columns[i],timeList[j], value))
print(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()), '---------------- Persistion finished!\n')
#result testing
print("---------- Inserstion Testing: ")
rows = session.execute("select * from factors_month where stock='600679.SH' and time > '2017-01-02' allow filtering")
for row in rows:
print(row.stock,row.factor,row.time,row.value)
# close connection with cassandra
cluster.shutdown()
# retrieve newly updated data
# monthRetrieve(datetime.date(2017,4,1), datetime.date(2017,4,30), fields1=['trade_status','close', 'mkt_freeshares','mkt_cap_float','roa'], multi_mfd = False)
monthRetrieve(datetime.date(2017,5,1), datetime.date(2017,5,31), fields1=['trade_status','close', 'mkt_freeshares'], multi_mfd = False)
# monthRetrieve(datetime.date(2017,1,1), datetime.date(2017,4,30), fields1=['roa'], multi_mfd = False)
# retrieve data from last year to 3.31 to inherit previous data
# monthRetrieve(datetime.date(2012,1,1), datetime.date(2017,4,30), fields1=['pe_ttm','pb_lf'], multi_mfd = False)