def insert_company(self):
print( "Insert new company information" )
mdb_query = Query()
iex = Iex()
#Get all symbols in MongoDB
mdb_symbols = mdb_query.get_symbols()
#Get companies already in MongoDB
mdb_companies = mdb_query.get_company( mdb_symbols['symbol'].tolist() )
#Initial call to print 0% progress
printProgressBar(0, len(mdb_symbols.index), prefix = 'Progress:', suffix = '', length = 50)
#Loop through symbols
for index, mdb_symbol in mdb_symbols.iterrows():
#Insert if no mdb company data exists
if mdb_companies.empty:
#Get company data from IEX
iex_company = iex.get_company( mdb_symbol["symbol"] )
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "Inserting company for " + mdb_symbol["symbol"] + " ", length = 50)
self.db.iex_company.insert_many( iex_company.to_dict('records') )
continue
#Skip company if already in MongoDB
if not mdb_companies[ mdb_companies['symbol'] == mdb_symbol['symbol'] ].empty:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "No new data for " + mdb_symbol["symbol"] + " ", length = 50)
continue
#Get company data from IEX
iex_company = iex.get_company( mdb_symbol["symbol"] )
#Insert if company data exists
if not iex_company.empty:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "Inserting company for " + mdb_symbol["symbol"] + " ", length = 50)
self.db.iex_company.insert_many( iex_company.to_dict('records') )
else:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "No data for " + mdb_symbol["symbol"] + " ", length = 50)
def insert_symbols(self):
print( "Insert new symbols" )
mdb_query = Query()
iex = Iex()
#Get all common stocks from IEX
symbols = iex.get_symbols(ref_type="cs")
#Get SPY (S&P500 exchange traded index) from IEX
symbols_spy = iex.get_symbols(ref_symbol="SPY")
#Reset indices (probably not necessary)
symbols.reset_index(drop=True, inplace=True)
symbols_spy.reset_index(drop=True, inplace=True)
#Append SPY to stocks
symbols = symbols.append(symbols_spy, ignore_index=True, sort=False)
symbols.reset_index(drop=True, inplace=True)
#Get symbols already in MongoDB
mdb_symbols = mdb_query.get_symbols()
#Initial call to print 0% progress
printProgressBar(0, len(symbols.index), prefix = 'Progress:', suffix = '', length = 50)
#Loop through symbols
for index, symbol in symbols.iterrows():
#Exclude forbidden characters
forbidden = ["#"]
if any( x in symbol["symbol"] for x in forbidden):
#Update progress bar
printProgressBar(index+1, len(symbols.index), prefix = 'Progress:', suffix = "Symbol contains forbidden character: " + symbol["symbol"] + " ", length = 50)
continue
#If MongoDB empty insert symbol
if mdb_symbols.empty:
#Update progress bar
printProgressBar(index+1, len(symbols.index), prefix = 'Progress:', suffix = "Inserting new symbol: " + symbol["symbol"] + " ", length = 50)
self.db.iex_symbols.insert_one( symbol.to_dict() )
else:
#Is symbol already in MongoDB
mask = (mdb_symbols['iexId'] == symbol['iexId']) & (mdb_symbols['isEnabled'] == symbol['isEnabled']) & (mdb_symbols['name'] == symbol['name']) & (mdb_symbols['type'] == symbol['type'])
#Insert if not in MongoDB
if mdb_symbols.loc[mask].empty:
#Update progress bar
printProgressBar(index+1, len(symbols.index), prefix = 'Progress:', suffix = "Inserting new symbol: " + symbol["symbol"] + " ", length = 50)
self.db.iex_symbols.insert_one( symbol.to_dict() )
else:
#Update progress bar
printProgressBar(index+1, len(symbols.index), prefix = 'Progress:', suffix = "Symbol " + symbol["symbol"] + " already exists ", length = 50)
def insert_quotes(self):
print( "Insert new quotes" )
mdb_query = Query()
iex = Iex()
#Get all symbols in MongoDB
mdb_symbols_full = mdb_query.get_active_companies()
#Get current date
currDate = datetime.datetime.now().strftime("%Y-%m-%d")
#Initial call to print 0% progress
printProgressBar(0, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = '', length = 50)
idx_min = 0
query_num = 100
while idx_min < len(mdb_symbols_full.index):
idx_max = idx_min + query_num
if idx_max > len(mdb_symbols_full.index):
idx_max = len(mdb_symbols_full.index)
mdb_symbols = mdb_symbols_full.iloc[ idx_min:idx_max ]
mdb_symbols.reset_index(drop=True, inplace=True)
#Get latest price in MongoDB for each symbol up to 50 days ago
mdb_quotes = mdb_query.get_quotes( mdb_symbols.tolist(), currDate, "latest" )
#Loop through symbols
iex_quotes = pandas.DataFrame()
for index, mdb_symbol in mdb_symbols.iteritems():
#Get matching quote in MongoDB
if not mdb_quotes.empty:
mdb_quote = mdb_quotes[ mdb_quotes['symbol'] == mdb_symbol ]
else:
mdb_quote = mdb_quotes
#continue if already up to date
if not mdb_quote.empty and (mdb_quote['date'].iloc[0] == currDate):
#Update progress bar
printProgressBar(idx_min+index+1, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = "No new data for " + mdb_symbol + " ", length = 50)
continue
#Get quote from IEX
iex_quote = iex.get_quote( mdb_symbol )
#Select quotes more recent than MongoDB
if not iex_quote.empty and not mdb_quote.empty:
mask = iex_quote['date'] > mdb_quote['date'].iloc[0]
iex_quote = iex_quote.loc[mask]
#Insert if quotes exist
if not iex_quote.empty:
#Update progress bar
printProgressBar(idx_min+index+1, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = "Inserting quote for " + mdb_symbol + " ", length = 50)
#Append quote
iex_quotes = iex_quotes.append(iex_quote, ignore_index=True, sort=False)
else:
#Update progress bar
printProgressBar(idx_min+index+1, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = "No new data for " + mdb_symbol + " ", length = 50)
#Bulk insert quotes
if not iex_quotes.empty:
self.db.iex_quotes.insert_many( iex_quotes.to_dict('records') )
idx_min = idx_min + query_num
def insert_financials(self):
print( "Insert new financials" )
mdb_query = Query()
iex = Iex()
#Get all symbols in MongoDB
mdb_symbols = mdb_query.get_active_companies()
#Get current date
currDate = datetime.datetime.now().strftime("%Y-%m-%d")
#Get latest financials in MongoDB for each symbol
mdb_financials = mdb_query.get_financials( mdb_symbols.tolist(), currDate, "latest" )
#Initial call to print 0% progress
printProgressBar(0, len(mdb_symbols.index), prefix = 'Progress:', suffix = '', length = 50)
#Loop through symbols
for index, mdb_symbol in mdb_symbols.iteritems():
#Get financials from IEX
iex_financials = iex.get_financials( mdb_symbol )
#Get matching financial in MongoDB
mdb_financial = mdb_financials[ mdb_financials['symbol'] == mdb_symbol ]
#Select financials more recent than MongoDB
if not mdb_financial.empty and not iex_financials.empty:
mask = iex_financials['reportDate'] > mdb_financial['reportDate'].iloc[0]
iex_financials = iex_financials.loc[mask]
#Insert if financials exist
if not iex_financials.empty:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "Inserting financials for " + mdb_symbol + " ", length = 50)
self.db.iex_financials.insert_many( iex_financials.to_dict('records') )
else:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "No new data for " + mdb_symbol + " ", length = 50)
def insert_balancesheets(self):
print( "Insert new balance sheets" )
mdb_query = Query()
iex = Iex()
#Get all symbols in MongoDB
mdb_symbols = mdb_query.get_active_companies()
#Get current date
currDate = datetime.datetime.now().strftime("%Y-%m-%d")
threeMonthsAgo = (pandas.Timestamp(currDate) + pandas.DateOffset(days=-120)).strftime('%Y-%m-%d')
#Get latest balancesheets in MongoDB for each symbol
mdb_balancesheets = mdb_query.get_balancesheets( mdb_symbols.tolist(), currDate, "latest" )
#Initial call to print 0% progress
printProgressBar(0, len(mdb_symbols.index), prefix = 'Progress:', suffix = '', length = 50)
#flag = False
#Loop through symbols
for index, mdb_symbol in mdb_symbols.iteritems():
#if mdb_symbol == 'YETI':
# flag = True
#if not flag:
# continue
#Get matching balancesheet in MongoDB
if not mdb_balancesheets.empty:
mdb_balancesheet = mdb_balancesheets[ mdb_balancesheets['symbol'] == mdb_symbol ]
else:
mdb_balancesheet = mdb_balancesheets
#Skip is less than 3 months since most recent
if not mdb_balancesheet.empty:
mask = mdb_balancesheet['reportDate'] > threeMonthsAgo
mdb_recent_balancesheet = mdb_balancesheet.loc[mask]
if not mdb_recent_balancesheet.empty:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "Data too recent for " + mdb_symbol + " ", length = 50)
continue
#Get balancesheets from IEX
iex_balancesheets = iex.get_balancesheets( mdb_symbol )
#Select balancesheets more recent than MongoDB
if not mdb_balancesheet.empty and not iex_balancesheets.empty:
mdb_balancesheet = mdb_balancesheet.sort_values(by='reportDate', ascending=False, axis='index')
mask = iex_balancesheets['reportDate'] > mdb_balancesheet['reportDate'].iloc[0]
iex_balancesheets = iex_balancesheets.loc[mask]
#Insert if balancesheets exist
if not iex_balancesheets.empty:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "Inserting balancesheets for " + mdb_symbol + " ", length = 50)
self.db.iex_balancesheets.insert_many( iex_balancesheets.to_dict('records') )
else:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "No new data for " + mdb_symbol + " ", length = 50)
def insert_dividends(self):
print( "Insert new dividends" )
mdb_query = Query()
iex = Iex()
#Get all symbols in MongoDB
#mdb_symbols = mdb_query.get_active_companies()
#Get current date
currDate = datetime.datetime.now().strftime("%Y-%m-%d")
#Get existing portfolios
portfolios = mdb_query.get_portfolios(currDate)[["portfolioID","inceptionDate"]]
#Loop through portfolios
mdb_symbols = pandas.DataFrame()
for portfolio_index, portfolio_row in portfolios.iterrows():
#Get portfolioID and inceptionDate
portfolio = portfolio_row['portfolioID']
inceptionDate = portfolio_row['inceptionDate']
#Default to calculating holdings from inception
date = inceptionDate
#Get current holdings table
holdings = mdb_query.get_holdings(portfolio, inceptionDate, "after")
#print( holdings )
mdb_symbols = mdb_symbols.append(holdings, ignore_index=True, sort=False)
#print( mdb_symbols )
mdb_symbols = mdb_symbols[mdb_symbols['symbol'] != 'USD']
mdb_symbols = mdb_symbols['symbol'].unique().tolist()
#print( mdb_symbols )
#quit()
#Get latest dividend in MongoDB for each symbol
mdb_dividends = mdb_query.get_dividends( mdb_symbols, currDate, "latest" )
#Initial call to print 0% progress
printProgressBar(0, len(mdb_symbols), prefix = 'Progress:', suffix = '', length = 50)
#flag = False
#Loop through symbols
for index, mdb_symbol in enumerate(mdb_symbols):
#if mdb_symbol["symbol"] == "ZZZZZZZZZ":
# flag = True
#if not flag:
# continue
#Get 1m of dividends from IEX
iex_dividends = iex.get_dividends( mdb_symbol, ref_range='1m' )
#Get matching dividend in MongoDB
mdb_dividend = mdb_dividends[ mdb_dividends['symbol'] == mdb_symbol ]
#Select dividends more recent than MongoDB
if not mdb_dividend.empty and not iex_dividends.empty:
mask = iex_dividends['exDate'] > mdb_dividend['exDate'].iloc[0]
iex_dividends = iex_dividends.loc[mask]
#Insert if dividends exist
if not iex_dividends.empty:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols), prefix = 'Progress:', suffix = "Inserting dividend for " + mdb_symbol + " ", length = 50)
#print( iex_dividends )
self.db.iex_dividends.insert_many( iex_dividends.to_dict('records') )
else:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols), prefix = 'Progress:', suffix = "No new data for " + mdb_symbol + " ", length = 50)
def insert_stats(self):
print( "Insert new stats" )
mdb_query = Query()
iex = Iex()
#Get all symbols in MongoDB
mdb_symbols_full = mdb_query.get_active_companies()
#mdb_symbols_full = mdb_symbols_full[mdb_symbols_full == 'A']
#Get current date
currDate = datetime.datetime.now().strftime("%Y-%m-%d")
#Initial call to print 0% progress
printProgressBar(0, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = '', length = 50)
idx_min = 0
query_num = 100
#flag = True
while idx_min < len(mdb_symbols_full.index):
#if idx_min > 1:
# break
idx_max = idx_min + query_num
if idx_max > len(mdb_symbols_full.index):
idx_max = len(mdb_symbols_full.index)
mdb_symbols = mdb_symbols_full.iloc[ idx_min:idx_max ]
mdb_symbols.reset_index(drop=True, inplace=True)
#Get latest price in MongoDB for each symbol up to 50 days ago
mdb_stats = mdb_query.get_stats( mdb_symbols.tolist(), currDate, "latest" )
#Loop through symbols
for index, mdb_symbol in mdb_symbols.iteritems():
#Get stat from IEX
#print( mdb_symbol )
#print( mdb_symbol["symbol"] )
#if mdb_symbol == "BAH":
# flag = False
#if flag:
# continue
iex_stat = iex.get_stats( mdb_symbol )
#Get matching stat in MongoDB
if not mdb_stats.empty:
mdb_stat = mdb_stats[ mdb_stats['symbol'] == mdb_symbol ]
else:
mdb_stat = mdb_stats
#Select stats more recent than MongoDB
if not iex_stat.empty and not mdb_stat.empty:
mask = iex_stat['date'] > mdb_stat['date'].iloc[0]
iex_stat = iex_stat.loc[mask]
#Insert if stats exist
#print( iex_stat )
if not iex_stat.empty:
#Update progress bar
printProgressBar(idx_min+index+1, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = "Inserting stat for " + mdb_symbol + " ", length = 50)
self.db.iex_stats.insert_many( iex_stat.to_dict('records') )
else:
#Update progress bar
printProgressBar(idx_min+index+1, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = "No new data for " + mdb_symbol + " ", length = 50)
idx_min = idx_min + query_num
def delete_duplicate_balancesheets(self, ref_date = "1990-01-01", when = "on"):
"""
Delete duplicates prices in MongoDB
@params:
ref_date - Optional : date YYYY-MM-DD (Str)
when - Optional : on, latest (Str)
"""
mdb_query = Query()
mdb_symbols = mdb_query.get_symbols()
#mdb_symbols = mdb_symbols.iloc[ 999: , : ]
mdb_symbols.reset_index(drop=True, inplace=True)
#mdb_symbols = ["A"]
#print( mdb_symbols )
printProgressBar(0, len(mdb_symbols.index), prefix = 'Progress:', suffix = '', length = 50)
for index, symbol in mdb_symbols.iterrows():
#if index > 10:
# break
#print( symbol )
query = { "symbol": { "$in": [symbol["symbol"]] },
"reportDate": { "$gte": ref_date } }
results = self.db.iex_balancesheets.find( query ).sort("reportDate", DESCENDING)
balancesheets = pandas.DataFrame()
for doc in results:
balancesheets = balancesheets.append( pandas.DataFrame.from_dict(doc, orient='index').T, ignore_index=True, sort=False )
duplicates = balancesheets[balancesheets.duplicated(['reportDate'])]
print( duplicates )
# Remove all duplicates in one go
#if not duplicates.empty:
# self.db.iex_charts.delete_many({"_id":{"$in":duplicates['_id'].tolist()}})
# Remove duplicates if they exist
if not duplicates.empty:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "Deleting duplicates for " + symbol["symbol"] + " ", length = 50)
#self.db.iex_balancesheets.delete_many({"_id":{"$in":duplicates['_id'].tolist()}})
else:
#Update progress bar
printProgressBar(index+1, len(mdb_symbols.index), prefix = 'Progress:', suffix = "No duplicates for " + symbol["symbol"] + " ", length = 50)
def insert_prices(self):
print( "Insert new charts" )
mdb_query = Query()
iex = Iex()
#Get all symbols in MongoDB
mdb_symbols_full = mdb_query.get_active_companies()
mdb_symbols_full = mdb_symbols_full[mdb_symbols_full == 'SPY']
#Get current date
currDate = datetime.datetime.now().strftime("%Y-%m-%d")
#Initial call to print 0% progress
printProgressBar(0, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = '', length = 50)
idx_min = 0
query_num = 100
#flag = False
while idx_min < len(mdb_symbols_full.index):
#if idx_min > 1:
# break
idx_max = idx_min + query_num
if idx_max > len(mdb_symbols_full.index):
idx_max = len(mdb_symbols_full.index)
mdb_symbols = mdb_symbols_full.iloc[ idx_min:idx_max ]
mdb_symbols.reset_index(drop=True, inplace=True)
#Get latest price in MongoDB for each symbol up to 50 days ago
mdb_charts = mdb_query.get_chart( mdb_symbols.tolist(), currDate, "latest" )
#print( mdb_charts )
#break
#Loop through symbols
for index, mdb_symbol in mdb_symbols.iteritems():
#Get matching chart in MongoDB
mdb_chart = mdb_charts[ mdb_charts['symbol'] == mdb_symbol ]
#Get first date to insert
date = mdb_chart.date.iloc[0]
date = (pandas.Timestamp(date) + pandas.DateOffset(days=1)).strftime('%Y-%m-%d')
iex_chart = pandas.DataFrame()
while date <= currDate:
#Get date in correct format for IEX query
iex_date = (pandas.Timestamp(date)).strftime('%Y%m%d')
#Get IEX chart
iex_chart_day = iex.get_chart( mdb_symbol, ref_range=iex_date )
iex_chart = iex_chart.append(iex_chart_day, ignore_index=True, sort=False)
#Increment date
date = (pandas.Timestamp(date) + pandas.DateOffset(days=1)).strftime('%Y-%m-%d')
#Get 1y of charts from IEX
#print( mdb_symbol )
#print( mdb_symbol["symbol"] )
#if mdb_symbol["symbol"] == "ZZZZZZZZZ":
# flag = True
#if not flag:
# continue
#iex_chart = iex.get_chart( mdb_symbol, ref_range='5d' )
#Get matching chart in MongoDB
#mdb_chart = mdb_charts[ mdb_charts['symbol'] == mdb_symbol ]
#Select charts more recent than MongoDB
#if not iex_chart.empty and not mdb_chart.empty:
# mask = iex_chart['date'] > mdb_chart['date'].iloc[0]
# iex_chart = iex_chart.loc[mask]
#Insert if charts exist
#print( iex_chart )
if not iex_chart.empty:
#Update progress bar
printProgressBar(idx_min+index+1, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = "Inserting chart for " + mdb_symbol + " ", length = 50)
#Print write error if couldn't insert charts
try:
print( iex_chart )
self.db.iex_charts.insert_many( iex_chart.to_dict('records') )
except BulkWriteError as bwe:
print( bwe.details )
raise
else:
#Update progress bar
printProgressBar(idx_min+index+1, len(mdb_symbols_full.index), prefix = 'Progress:', suffix = "No new data for " + mdb_symbol + " ", length = 50)
idx_min = idx_min + query_num