-
Notifications
You must be signed in to change notification settings - Fork 1
/
get_data
328 lines (268 loc) · 11.8 KB
/
get_data
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
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
# -------------------------------------------------------
# Importing Libraries
# -------------------------------------------------------
import tpqoa
import sqlite3
import pandas as pd
import datetime as dt
import time
from datetime import datetime
from sqlite3 import Error
from tqdm.auto import tqdm
# -------------------------------------------------------
# -------------------------------------------------------
# SQL Database
# -------------------------------------------------------
class SQLDB:
def __init__(self, db_name: str):
"""
:param db_name: The name of the database to create or access.
"""
self.db_name = db_name
try:
self.con = sqlite3.connect(self.db_name)
print("Connection established: " + self.db_name)
except Error:
print("Connection failed: " + self.db_name)
print(Error)
self.cursor = self.con.cursor()
try:
print('\n'+"Tables Available to Query from " + self.db_name + ":")
print(self.cursor.execute("""SELECT name FROM sqlite_master WHERE type='table';""").fetchall(), '\n')
except sqlite3.Error as error:
print("Failed to execute the above query", error)
def df_to_table(self, df: pd.DataFrame or pd.Series, table_name: str):
"""
:param df: pd.DataFrame to save into SQL database.
:param table_name: name to save table in SQL database.
:return: SQL database saved to directory.
"""
print("Adding " + table_name + " in " + self.db_name)
df.to_sql(name=table_name, con=self.con, if_exists='replace')
self.con.commit()
self.con.close()
print("Connection closed: " + self.db_name, '\n')
def update_db(self, df: pd.DataFrame or pd.Series, table_name: str):
"""
:param df: pd.DataFrame to update existing SQL table.
:param table_name: name to save table in SQL database.
:return: SQL database updated with new files.
"""
print("Updating " + table_name + " in " + self.db_name)
df.to_sql(name=table_name, con=self.con, if_exists='append')
self.con.commit()
self.con.close()
print("Connection closed: " + self.db_name, '\n')
def table_to_df(self, table_name: str):
"""
:param table_name: name of table in SQL database.
:return: pd.DataFrame of data from table in SQL db.
"""
print('Importing ' + table_name + ' From ' + self.db_name)
df = pd.read_sql("SELECT * FROM " + table_name, self.con)
self.con.close()
print("Connection Closed: " + self.db_name, '\n')
return df
def query_db(self, query: str):
"""
:param query: an sql query in string format
:return: an SQL query into a pd.DataFrame.
"""
try:
return pd.read_sql_query(query, self.con)
except Error:
print("Failed to execute the above query", Error)
# -------------------------------------------------------
# -------------------------------------------------------
# API Call Class
# -------------------------------------------------------
class APIData:
def __init__(self, granularity, start_date):
"""
Sets the start date, end date and candle frequency to download.
"""
# Start Date
self.start_date = datetime.strptime(start_date, '%Y-%m-%d')
# End Date
self.end_date = datetime.now()
# Candle Frequency
self.granularity = granularity
print("Model Frequency = " + self.granularity + '\n')
print('<<<<<< Calling API >>>>>>')
self.api = tpqoa.tpqoa("oanda.cfg")
# print(self.api.account_type, " | ", self.api.account_id)
# List the available instruments and save to db
print('\n'+'###### Calling SQL DB ######')
self.inst = pd.DataFrame(self.api.get_instruments()).pop(1)
SQLDB('Forex_db').df_to_table(self.inst, 'instruments')
# Setting the maximum look-back period based on the frequency
max_candle = None
if self.granularity == 'D':
max_candle = 5000
if self.granularity == 'H12':
max_candle = 2500
if self.granularity == 'H8':
max_candle = 1658
if self.granularity == 'H6':
max_candle = 1250
if self.granularity == 'H4':
max_candle = 832
if self.granularity == 'H3':
max_candle = 625
if self.granularity == 'H2':
max_candle = 416
if self.granularity == 'H1':
max_candle = 208
if self.granularity == 'M30':
max_candle = 208
if self.granularity == 'M15':
max_candle = 208
if self.granularity == 'M10':
max_candle = 208
if self.granularity == 'M5':
max_candle = 208
if self.granularity == 'M1':
max_candle = 208
# api maximum period for this frequency:
delta = dt.timedelta(days=max_candle)
# Iterating from start date, recording date ranges of days
self.date_ranges = []
temp_start_date = self.start_date
while temp_start_date < self.end_date:
temp_end_date = temp_start_date + delta
if temp_end_date > self.end_date:
temp_end_date = self.end_date
self.date_ranges.append([temp_start_date, temp_end_date])
temp_start_date = temp_end_date + dt.timedelta(days=1)
def get_instrument(self, ticker: str, price: str):
"""
:param ticker: string of instrument EUR_USD, GBP_JPY
:param price: string of 'A' = ask price; 'B' = bid price
:return: pd.DateFrame of OHLCV instrument data.
"""
s_price = None
if price == 'A':
s_price = '_Ask'
elif price == 'B':
s_price = '_Bid'
print('Downloading ' + ticker + s_price)
tables = []
# For each date range, pass dates into API
for start_dt, end_dt in tqdm(self.date_ranges):
start = start_dt.strftime("%Y-%m-%dT00:00")
end = end_dt.strftime("%Y-%m-%dT%H")
try:
item = self.api.get_history(instrument=ticker,
start=start,
end=end,
granularity=self.granularity,
price=price)
tables.append(item)
time.sleep(1)
except Exception as exception_:
print("An Error has occurred: ", exception_)
pass
single_df = pd.concat(tables)
# Format and set column index's and name
single_df.drop('complete', axis=1, inplace=True)
single_df.reset_index(inplace=True)
single_df.rename(
columns={'time': 'DateTime',
'o': ticker + '_' + 'Open',
'h': ticker + '_' + 'High',
'l': ticker + '_' + 'Low',
'c': ticker + '_' + 'Close',
'volume': ticker + '_' + 'Volume'}, inplace=True)
single_df.set_index(['DateTime'], inplace=True)
# Puts instrument name as table header in another level
# single_df.columns = pd.MultiIndex.from_product([[ticker], single_df.columns])
return single_df
def all_instruments(self):
"""
Downloads all OANDA instruments and places into a bid and ask dataframe.
"""
print('Searching from: ', self.start_date, ' | ', self.end_date, ' at a frequency of:', self.granularity)
print('<<<<<< Downloading Instrument Database >>>>>>')
ask = []
bid = []
for i in range(len(self.inst)):
askdf = self.get_instrument(self.inst[i], "A")
ask.append(askdf)
biddf = self.get_instrument(self.inst[i], "B")
bid.append(biddf)
ask = pd.concat(ask, axis=1)
bid = pd.concat(bid, axis=1)
print('<<<<<< Download Complete >>>>>>')
return ask, bid
def __update_instrument(self, df: pd.DataFrame, ticker: str, price: str):
"""
Update database with current OHLCV.
:param df: Existing pd.DataFrame with OHLCV to update.
:param ticker: string of instrument EUR_USD, GBP_JPY.
:param price: string of 'A' = ask price; 'B' = bid price.
:return: pd.DateFrame of updated OHLCV instrument data.
"""
s_price = None
if price == 'A':
s_price = '_Ask'
elif price == 'B':
s_price = '_Bid'
# Check using existing database and current datetime
dt_now = dt.datetime.strptime(datetime.now().strftime("%Y-%m-%d %H:00"), '%Y-%m-%d %H:00')
dt_df = dt.datetime.strptime(df.index[-1], '%Y-%m-%d %H:00:00')
check = dt_now - dt_df
if check > dt.timedelta(hours=df.index[1]-df.index[0]):
try:
print('Updating ' + ticker + s_price)
df_update = self.api.get_history(instrument=ticker,
start=df.index[-1],
end=datetime.now().strftime("%Y-%m-%d %H:00"),
granularity=self.granularity,
price=price)
# Format and set column index's and name
df_update.drop('complete', axis=1, inplace=True)
df_update.reset_index(inplace=True)
df_update.rename(columns={'time': 'DateTime',
'o': ticker + '_' + 'Open',
'h': ticker + '_' + 'High',
'l': ticker + '_' + 'Low',
'c': ticker + '_' + 'Close',
'volume': ticker + '_' + 'Volume'}, inplace=True)
df_update.set_index(['DateTime'], inplace=True)
return df_update.drop(df_update.index[0])
except Exception as exception_:
print("An Error has occurred: ", exception_)
else:
print('No Updates to be Downloaded for: ' + ticker + s_price)
pass
def update_instruments(self, df: pd.DataFrame):
"""
Update database to most recent datetime.
:param df: Existing pd.DataFrame with OHLCV to update.
:return: pd.DateFrame of every instrument OHLCV data updated.
"""
print('<<<<<< Updating Instrument Database >>>>>>')
ask = []
bid = []
for i in tqdm(range(len(self.inst))):
askdf = self.__update_instrument(df, self.inst[i], "A")
ask.append(askdf)
biddf = self.__update_instrument(df, self.inst[i], "B")
bid.append(biddf)
try:
ask = pd.concat(ask, axis=1)
bid = pd.concat(bid, axis=1)
except: pass
print('<<<<<< Update Complete >>>>>>')
return ask, bid
# -------------------------------------------------------
if __name__ == '__main__':
model_1D = APIData('D', '2010-1-1')
# Download all instruments
ask_1D, bid_1D = model_1D.all_instruments()
SQLDB('Forex_db').df_to_table(ask_1D, 'D1_todo_market_ask')
SQLDB('Forex_db').df_to_table(bid_1D, 'D1_todo_market_bid')
# Update existing database
ask_1D_update, bid_1D_update = model_1D.update_instruments(ask_1D)
SQLDB('Forex_db').update_db(ask_1D_update, 'D1_todo_market_ask')
SQLDB('Forex_db').update_db(bid_1D_update, 'D1_todo_market_bid')