'''
Created on Dec 16, 2014
@author: Brett Paufler
Copyright Brett Paufler

'''


import pandas as pd
import os
import sqlite3
from pandas.io import sql


pd.set_option('expand_frame_repr',False)



tdays = {0:"Monday",
         1:"Tuesday",
         2:"Wednesday",
         3:"Thursday",
         4:"Friday"}

prefix = {"corn":"CZ",
          "soybeans":"SX"}

weekdayList = ["Monday","Tuesday","Wednesday","Thursday","Friday"]

cM = {"corn": "./corn/",
      "soybeans":"./soybean/",
     }

graphList = [["corn",2014],
             ["corn",range(1960,2015)],
             ["soybeans",2014],
             ["soybeans",range(1970,2015)]
             ]


def report(dF):
    '''
    Quick Debugging Screen Output to visualize current status of dataFrame
    '''
    print "Quick Report %s" % str(dF.shape)
    print dF.head()
    

def get_trades(commodity):
    '''
    returns all trades of given commodity,
        no filter for year
        if only want trades for year in which issue expires,
            see by_issue_year
    '''


    
    if commodity not in cM:
        raise Exception ("%s Not a Valid Commodity Type (check spelling, mapping)")
    

    tA = []
    for f in os.listdir(cM[commodity]):
        print "Extracting Data from %s" % f

        if commodity == "corn":
            issueName = f[f.find("CZ"):]
            issueName = issueName[:6]
        if commodity == "soybeans":
            issueName = f[f.find("SX"):]
            issueName = issueName[:6]
        
        if not issueName:
            raise Exception("Error defining issueName for %s" % commodity)
        
        
        t = pd.read_csv(cM[commodity] + f, parse_dates=['Date'])
        t["Issue"] = issueName
        
        tA.append(t)

    dF = pd.concat(tA)
    print "\n\n"
    print "%s returned from get_trades: shape %s" % (commodity,str(dF.shape))
    print dF.head(2)
    print "..."
    print dF.tail(2)
    return dF


def by_issue_year(commodity="corn", years=range(1960,2016)):
    '''
    returns a dataFrame in which the Issue and Year are the same
    
        by_year("corn", 2015)
            returns all dates in 2015 for CZ2015
        by_year("soybeans", [2013,2014])
            returns all dates in 2013 for SX2013 and in 2014 for SX2014        
        by_year(commodity="corn", years=range(1960,2016))
            returns all corn trades from 1960 to 2015 for the year in which the issue comes due
    '''
    
    if isinstance(years, int):
        years = [years]
    
    dF = get_trades(commodity)
    
    temp = []
    for year in years:
        t = dF[dF.Issue == "%s%d" % (prefix[commodity],year)]
        #trades.report(t)
        t = t[t.Date > "%d-1-1" % year]
        #trades.report(t)
        t = t[t.Date < "%d-12-31" % year]
        #trades.report(t)
        temp.append(t)
    dF = pd.concat(temp)
    dF = dF.set_index(dF["Date"], verify_integrity=True)
    #report(dF)
    return dF



if __name__ == '__main__':

    dF = get_trades('soybeans')
    print dF.head()




'''

#Find Specific Issue
dF = dF[dF["Issue"].isin(["SX2015"])]

#Find Series of Issues
dF = dF[dF["Issue"].isin(["SX2014","SX2015"])]

#Twenty Day Moving Average
dF['Moving Average'] = pd.stats.moments.rolling_mean(dF['Settle'], window=20)


print dF.head()
print dF.tail()
print dF.shape


#simplest graph
dF.plot()
plt.show()
plt.savefig(outputName)

'''