class streamer(object):
def __init__(self):
"""
Reads in credentials from the user's .netrc file and creates
a Datastream object to be used in fetching data.
"""
rc = netrc()
uname, account, passwd = rc.authenticators('datastream')
self.DWE = Datastream(username=uname, password=passwd)
def _fetch_individual_code(self, code, fields=None, **kwargs):
"""
Internal function to fetch the data.
It offers two advantages to the original:
(1) It raises a warning, not an Exception when it has problems.
This is primarily useful when you have a list of codes and
want to know when one fails, but move on.
(2) It renames the colums so that they have the code as a part
of the column name, not just the field. This
"""
print(code) #TEMP
try:
df = self.DWE.fetch(code, fields=fields, **kwargs)
except Exception, e:
df = None
warnings.warn(("Unable to load {} \n".format(code)
+ str(e.message) + '\n'))
else:
def __init__(self):
"""
Reads in credentials from the user's .netrc file and creates
a Datastream object to be used in fetching data.
"""
rc = netrc()
uname, account, passwd = rc.authenticators('datastream')
self.DWE = Datastream(username=uname, password=passwd)
import pydatastream
from pydatastream import Datastream
import os
import csv
credentials = {}
with open('credentials.txt', 'r') as credentials_file:
line = credentials_file.readline().split(',')
credentials['username'] = line[0]
credentials['password'] = line[1]
DWE = Datastream(username=credentials['username'], password=credentials['password'])
DWE.raise_on_error = False
INACTIVE_DATE = ['WC07015']
DATATYPES = ['P', 'MV', 'PE', 'VA', 'VO', 'PTBV', 'MTBV']
shenzhen_equities_symbols = []
with open('shenzhen-equities-rmb.csv', 'r') as shenzhen_equities_csv:
reader = csv.DictReader(shenzhen_equities_csv)
shenzhen_equities_symbols = [line['Symbol'] for line in reader]
for stock_mnem in shenzhen_equities_symbols:
filename = 'stock_%s_%s.csv' % ('shenzhen', stock_mnem)
filename = filename.replace(':', '-')
if os.path.isfile('csvs/%s' % filename):
print 'Stock data already exists for %s' % (stock_mnem)
continue
import pandas as pd
from datetime import datetime
from pydatastream import Datastream
from utils import write_pickle
from private import DS_USERNAME, DS_PASSWORD
DWE = Datastream(username=DS_USERNAME, password=DS_PASSWORD)
def get_prices(mnem, date_from, date_to):
template = '{}(RI)~~USD'.format(mnem)
data = DWE.fetch(template, date_from=date_from, date_to=date_to)
return data['P'].rename(mnem)
holdings = pd.read_csv('assets/holdings.csv')
date_from = datetime(2016, 12, 31)
date_to = datetime(2017, 12, 31)
prices = []
exceptions = []
for i, row in holdings.iterrows():
try:
print(i / len(holdings))
prices.append(get_prices(row['ISIN'], date_from, date_to))
except Exception as error:
print(error)
print(row['ISIN'])
exceptions.append(row['ISIN'])
print('Re-download exceptions')
for isin in exceptions:
__author__ = 'gio'
from sqlalchemy import create_engine
engine = create_engine('postgresql://localhost/gio')
import matplotlib.pyplot as plt
from pydatastream import Datastream
DWE = Datastream(username="******", password="******")
data = DWE.get_price('@AAPL', date_from='2008', date_to='2016')
print data.head()
plt.plot(data['P'])
plt.ylabel('Apple')
plt.show()
def DatastreamInit():
from pydatastream import Datastream
key = getkey("datastream")
return Datastream(username=key.user, password=key.password)
import pg8000
from pydatastream import Datastream
DWE = Datastream(username="******", password="******")
conn = pg8000.connect(user="******", password="")
cursor = conn.cursor()
cursor.execute("SELECT isin FROM TRIAL1")
results = cursor.fetchall()
print(results)
b = type(results)
print (b)
for item in results:
print (item)
DWE.raise_on_error = False
data = DWE.get_price(item, date_from='2008', date_to='2016')
a = type (data)
print (a)
print(data)
if numOfArgs != 3:
print "Please run this python script with username,password and input file location in that order respectively."
exit()
#Setup login credentials and input file location
username = str(sys.argv[1])
pw = str(sys.argv[2])
input_file_loc = dir_input + str(sys.argv[3])
#Ensure that the input file location exists
if ( not os.path.isfile(str(input_file_loc)) ):
print "The file " + str(input_file_loc) + " does not exist."
exit()
#login credentials to datastream
DWE = Datastream(username=username,password=pw)
#other info from datastream
info = DWE.system_info()
subscribed_sources = DWE.sources()
#replace missing data with NaNs
DWE.raise_on_error = False
#get all codes, groups, start dates from input file
with open(input_file_loc,'r') as input_file:
symbol_ref = json.load(input_file)
#download timestamp
download_date = {'Custom_Download_Date' : datetime.datetime.now().isoformat()}
#calculate time taken for entire process
