-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathget_extraData.py
171 lines (129 loc) · 5.43 KB
/
get_extraData.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
import pandas as pd
import io
import requests
import os
import time
import numpy as np
from datetime import datetime
from dateutil.relativedelta import relativedelta
def transform_date(date): # 民國轉西元
y, m, d = date.split("/")
return str(int(y) + 1911) + "/" + m + "/" + d
def process_data(data):
data = data.replace(",", "")
data = data.replace("--", "")
if data == "":
return np.nan
return data
def gen_iter_date_by_month(start, end):
while end >= start:
yield start
start = start + relativedelta(months=1)
def getDatas(path):
datas = {}
for dirPath, dirNames, fileNames in os.walk(path):
if len(fileNames) == 0:
continue
fileNames.sort()
# 移除當月資料，以免資料不全
current = f"{(datetime.now() + relativedelta(day=1)).strftime('%Y%m%d')}.csv"
if fileNames[-1] == current:
os.remove(os.path.join(dirPath, fileNames[-1]))
del fileNames[-1]
print("remove", fileNames[-1])
for f in fileNames:
datas[os.path.splitext(f)[0]] = True
return datas
def save_twse_ftse_index(s, symbol, url_symbol, start):
savePath = os.path.join("./extraData", symbol)
os.makedirs(savePath, exist_ok=True)
datas = getDatas(savePath)
end = datetime.now() + relativedelta(day=1) # 設定為當月的 1 號
for day in gen_iter_date_by_month(start, end):
d = day.strftime("%Y%m%d")
if datas.get(d, False):
print(symbol, d, "already exists")
continue
url = f"https://www.twse.com.tw/rwd/zh/FTSE/{url_symbol}?response=csv&date={d}"
print(symbol, url_symbol, d, "saving...", f"from {url}")
c = s.get(url).content
try:
df = pd.read_csv(io.StringIO(c.decode("big5")), header=1)
print("raw data")
print(df)
except pd.errors.EmptyDataError:
print(symbol, d, "is empty")
return
print("after drop")
df = df.dropna(axis=1)
df = df[df[symbol.replace("指數", "報酬指數")] != "--"]
print(df)
df.loc[:, "Date"] = pd.to_datetime(df["日期"].apply(transform_date), format="%Y/%m/%d")
df.loc[:, "Close"] = df[symbol].apply(process_data).astype(float)
df.loc[:, "Adj Close"] = (
df[symbol.replace("指數", "報酬指數")].apply(process_data).astype(float)
)
df.loc[:, "Dividends"] = 0
df.loc[:, "Stock Splits"] = 0
time.sleep(5)
df = df[df["Adj Close"] != 0]
df.to_csv(os.path.join(savePath, f"{d}.csv"), index=False, lineterminator="\n")
def save_TAI50I_index(s):
save_twse_ftse_index(s, "臺灣50指數", "TAI50I", start=datetime(2002, 10, 1))
def save_TAI100I_index(s):
save_twse_ftse_index(s, "臺灣中型100指數", "TAI100I", start=datetime(2004, 11, 1))
def save_TAIDIVIDI_index(s):
save_twse_ftse_index(s, "臺灣高股息指數", "TAIDIVIDI", start=datetime(2007, 1, 1))
def save_TAIEX_index(s):
symbol = "臺灣加權股價指數"
savePath = os.path.join("./extraData", symbol)
os.makedirs(savePath, exist_ok=True)
datas = getDatas(savePath)
start = datetime(2003, 1, 1)
end = datetime.now() + relativedelta(day=1) # 設定為當月的 1 號
for day in gen_iter_date_by_month(start, end):
d = day.strftime("%Y%m%d")
if datas.get(d, False):
print(symbol, d, "already exists")
continue
histURL = f"https://www.twse.com.tw/rwd/zh/TAIEX/MI_5MINS_HIST?response=csv&date={d}"
totalReturnURL = f"https://www.twse.com.tw/rwd/zh/TAIEX/MFI94U?response=csv&date={d}"
print(symbol, d, "get history...", f"from {histURL}")
c = s.get(histURL).content
try:
hist = pd.read_csv(io.StringIO(c.decode("big5")), header=1)
print("hist raw data")
print(hist)
print("after drop")
hist = hist.dropna(axis=1)
print(hist)
except pd.errors.EmptyDataError:
print(symbol, d, "is empty")
return
print(symbol, d, "get total return...", f"from {totalReturnURL}")
c = s.get(totalReturnURL).content
totalReturn = pd.read_csv(io.StringIO(c.decode("big5")), header=1)
print("totalReturn raw data")
print(totalReturn)
print("after drop")
totalReturn = totalReturn.dropna(axis=1)
print(totalReturn)
df = hist.join(totalReturn)
assert df[df["日期"] != df["日 期"]].empty
df.loc[:, "Date"] = pd.to_datetime(df["日期"].apply(transform_date), format="%Y/%m/%d")
df.loc[:, "Open"] = df["開盤指數"].apply(process_data).astype(float)
df.loc[:, "High"] = df["最高指數"].apply(process_data).astype(float)
df.loc[:, "Low"] = df["最低指數"].apply(process_data).astype(float)
df.loc[:, "Close"] = df["收盤指數"].apply(process_data).astype(float)
df.loc[:, "Adj Close"] = df["發行量加權股價報酬指數"].apply(process_data).astype(float)
df.loc[:, "Dividends"] = 0
df.loc[:, "Stock Splits"] = 0
time.sleep(5)
df = df[df["Adj Close"] != 0]
df.to_csv(os.path.join(savePath, f"{d}.csv"), index=False, lineterminator="\n")
if __name__ == "__main__":
with requests.Session() as s:
save_TAI50I_index(s)
save_TAI100I_index(s)
save_TAIEX_index(s)
save_TAIDIVIDI_index(s)