yapywrangler
Update 05/31/2019:
This package was initally released over a year ago in the wake of Yahoo Finance updating their backend API.
At that time, the only means of retrieving data required gnarly parsing and took noticable time to process.
Because of this, on the initial release, a local file system for caching parts of retrieved files was used
to cut down on processing time.
Since then, Yahoo Finance has updated their API resulting in much faster response times, and negating the
need for the caching system.
Due to this, yapywrangler has been revised to include just its core request functionality.
Install:
pip install yapywrangler
Usage:
from yapywrangler import get_yahoo_data
stock_ticker = "MSFT"
data = get_yahoo_data(stock_ticker, start_date='2006-03-26', end_date='2019-05-29')
msft = pd.DataFrame(data)
Default Args
start_date: 2000-01-01
end_date: now
Data Format:
{
'timestamp': [1559136600, 1559223000, 1559321620],
'date': ['2019-05-29', '2019-05-30', '2019-05-31'],
'open': [29.0, 28.399999618530273, 27.56999969482422],
'high': [29.31999969482422, 28.559999465942383, 28.360000610351562],
'low': [27.729999542236328, 27.600000381469727, 27.5],
'close': [28.09000015258789, 28.030000686645508, 27.940000534057617],
'volume': [99969600, 65072900, 40126715]
}