fun with stock quotes

[Robin Becker]

In article <Fnn296.A3w at world.std.com>, Will Ware <wware at world.std.com> writes
>I recently discovered that Yahoo's stock-quote-serving web
>page is now offering historical stock data in machine-friendly
>form. The following scripts will pull in a bunch of stock data
>and format it as a C module. Then you can back-test any silly
>investment algorithms you come up with. Nerdy greedy fun for
>the whole family.
>
here's my take on that which just returns a class with all the values
######################################################
import urllib, string, normalDate, math, regex, sys, os

class _YahooData:
    def __init__(self,ticker,start,finish,Open,Low,High,Close,Volume):
        self.ticker=ticker
        self.start=start
        self.finish=finish
        self.Open=Open
        self.Low=Low
        self.High=High
        self.Close=Close
        self.Volume=Volume

def fetchRangeAsCSV(ticker, first, last):
    '''go to Yahoo and fetch a range of historical stock
    data, where 'first' and 'last' are normalDates'''
    key = (first, last)
    url = 'http://chart.yahoo.com/table.csv?s=%s' \
          '&a=%d&b=%d&c=%d&d=%d&e=%d&f=%d&g=d&q=q&y=0&z=%s' \
          '&x=.csv' % (ticker,
                       first.month(), first.day(), first.year(),
                       last.month(), last.day(), last.year(),
                       ticker)
    return urllib.URLopener().open(url).readlines()

def _yahoo_to_normalDate(date):
    date = string.split(date, '-')
    date[0] = eval(date[0])
    date[2] = eval(date[2])
    if date[2] > 90: year = date[2] + 1900
    else: year = date[2] + 2000
    month = ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep',
            'Oct','Nov','Dec'].index(date[1]) + 1
    return normalDate.ND(year * 10000 + month * 100 + date[0])

def fetchRange(ticker, first, last):
    Open = []
    Low = []
    High = []
    Close = []
    Volume = []
    finish=None
    for x in fetchRangeAsCSV(ticker,first,last)[1:]:
        F = string.split(x, ',')
        Open.append(eval(F[1]))
        Low.append(eval(F[2]))
        High.append(eval(F[3]))
        Close.append(eval(F[4]))
        Volume.append(eval(F[5]))
        if finish is None:
            finish=_yahoo_to_normalDate(F[0])

    start = _yahoo_to_normalDate(F[0])

    # Yahoo gives reverse time order
    Open.reverse()
    Low.reverse()
    High.reverse()
    Close.reverse()
    Volume.reverse()
    if len(data) == 0:
        raise 'ouch'
    return _YahooData(ticker,start,finish,Open,Low,High,Close,Volume)

if __name__=='__main__':
    # just do a quick test for now
    masterList = [
        # tech stocks
        'LNUX', 'MSFT', 'AMZN', 'INTC', 'RHAT', 'DELL', 'GTW',
        'CSCO', 'AAPL', 'AOL', 'COMS', 'CPQ', 'LCOS', 'LU',
        'ORCL', 'SUNW', 'TXN', 'YHOO',
        # single-letters and blue chips
        'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H',
        'J', 'K', 'L', 'N', 'O', 'P', 'R',
        'U', 'W', 'X', 'Y', 'Z',
        # Dow Jones stocks
        'AA', 'AXP', 'T', 'BA', 'CAT', 'CHV', 'KO',
        'DIS', 'DD', 'EK', 'GE', 'GM', 'GT', 'HWP',
        'IBM', 'IP', 'JNJ', 'MCD', 'MRK', 'MMM', 'JPM',
        'MO', 'PG', 'S', 'TRV', 'UK', 'WMT',
        # biotech
        'AFFX', 'BIOM', 'CRA', 'ENMD', 'HGSI',
        'MATX', 'MLNM', 'RGEN', 'SAFS', 'DNA', 'GENE',
        'INCY', 'GLGC', 'PFE',
        'AVXT', 'CTIC', 'CRXA', 'GZMO', 'IDPH',
        'ILXO', 'ISIP', 'RZYM', 'SUPG', 'TGEN', 'VICL',
        # retailers, general interest
        'BBY', 'COST', 'UPS', 'HD',
        ]
    masterList = masterList[:12]
    start = normalDate.ND(19950101)
    finish = normalDate.ND()   # today
    d = {}

    for t in masterList:
        d[t]=fetchRange(t, start, finish)
        print   "data for %s obtained start=%d(%f) finish=%d(%f)" % \
                (t,d[t].start,d[t].Close[0],d[t].finish,d[t].Close[-1])
-- 
Robin Becker