Script_FormatCSV.py

Go to the documentation of this file.

#!/usr/bin/python

import sys
import string

import datetime

# correct header, better to do
# correct date
# to do more use of wrapper

def printUsage():
   print '''
   ===================================================================
   USAGE
    ./Script_FormatCSV input_file FORMAT_input FORMAT_output > output_file

    FORMAT: CSV_Nyse /implemented
            CSV_Yahoo /implemented
            MyCSV /by default, not implemented
            MydCSV
            dCSV_Bourso /not implemented, no use just modify output in Server
                        / to modify output of Server, temporary

   ===================================================================
   '''

class CSV:

    def __init__(self,format_in,format_out,file):

        self.ListCSV=[]

        self.format_in=format_in
        self.format_out=format_out
        self.file=file
        

    def write(self):

        #print "# 5"
        # #0.1 DAY   2012-10-10 2012-10-22
        # To use wrapper C++ for header ?
        print "#0.1 DAY   "

        for i in range(len(self.ListCSV)):
            
            dictCSV = self.ListCSV[i]
            #print "dictCSV ",dictCSV
            if self.format_out == "MyCSV":
                print "%s %s,%.3f,%.3f,%.3f,%.3f,%i" %(dictCSV['date'],dictCSV['time'],dictCSV['open'],dictCSV['high'],dictCSV['low'],dictCSV['close'],dictCSV['volume'])
                
            elif self.format_out == "MydCSV":
                print "%s %s,%.3f,%i" %(dictCSV['date'],dictCSV['time'],dictCSV['value'],dictCSV['volume'])
                

    def read(self):

        
        status_data = False

        for line in self.file.xreadlines():

            dictCSV={}

            ##Read comment
            if status_data == False:

                
                if self.format_in == "CSV_Nyse" :
            
                    ret = string.find(line,"Date;opening;High;Low;closing;Volume")
                    #print "ret ", ret
                    #if ret ==0 :
                    #    status_data = True

                if self.format_in == "CSV_Yahoo":
                    ret =  string.find(line,"Date,Open,High,Low,Close,Volume,Adj Close")
                    #print "ret ", ret
                    
                if self.format_in == "dCSV_Bourso":
                    ret = string.find(line,"#")
                    #print ret

                    
                if ret == 0 :
                    status_data = True

            #status_data is True
            else:
                               
                if self.format_in == "CSV_Nyse" :

                    tabline=line.split(";")

                    #extract datetime
                    ## strptime is a static method of the datetime class
                    date_tmp=datetime.datetime.strptime( tabline[0], "%d/%m/%y" )
                    
                    dictCSV['date'] = date_tmp.strftime("%d-%m-%Y")
                    dictCSV['time']="00:00:00"
                    dictCSV['open']=float(tabline[1])
                    dictCSV['high']=float(tabline[2])
                    dictCSV['low']=float(tabline[3])
                    dictCSV['close']=float(tabline[4])
                    dictCSV['volume']=int(tabline[5])

                if self.format_in == "CSV_Yahoo" :

                    tabline=line.split(",")

                    #extract datetime
                    ## strptime is a static method of the datetime class
                    #print "tabline[0] ", tabline[0]
                    date_tmp=datetime.datetime.strptime( tabline[0], "%Y-%m-%d" )
                   
                    # original
                    #dictCSV['date'] = date_tmp.strftime("%d-%m-%Y")
                    # more recent, do not change format, could copy or test correct parsing
                    dictCSV['date']=date_tmp.strftime("%Y-%m-%d")
                    dictCSV['time']="00:00:00"
                    dictCSV['open']=float(tabline[1])
                    dictCSV['high']=float(tabline[2])
                    dictCSV['low']=float(tabline[3])
                    dictCSV['close']=float(tabline[4])
                    dictCSV['volume']=int(tabline[5])
                    # adjusted close would be 6
                    
                if self.format_in == "dCSV_Bourso" :
                    tabline=line.split(",")
                    #print "tabline ",tabline
                    #date_tmp=datetime.datetime.strptime( tabline[0], "%d/%m/%y %H:%M:%S" )
                    date_tmp=datetime.datetime.strptime( tabline[0], "%d-%m-%y %H:%M:%S" ) 
                    dictCSV['date'] = date_tmp.strftime("%Y-%m-%d")
                    #dictCSV['time']="00:00:00"
                    dictCSV['time'] = date_tmp.strftime("%H:%M:%S")
                    dictCSV['value'] = float(tabline[1])
                    dictCSV['volume'] = int(round(float(tabline[2]),0))
                    
                    
                self.ListCSV.append(dictCSV)
                #print self.ListCSV

        ##end reading
        self.file.close()

        ##want reverse of chronologic order
        if self.format_in == "CSV_Nyse":
            self.ListCSV.reverse()

###end class CSV              
                                             


####### main
ListFormat= [ "CSV_Nyse", "CSV_Yahoo", "dCSV_Bourso", "MyCSV", "MydCSV" ]

try:
    Infile = open(sys.argv[1],'r')
    Format_in = sys.argv[2]
    Format_out = sys.argv[3]
except IndexError:
    printUsage()
    sys.exit(1)

#print "infile ",Infile
#print "Format_in ", Format_in
#print "Format_out ", Format_out

if ( ( Format_in not in ListFormat ) or ( Format_out not in ListFormat) ):
    print "Error Format"
    printUsage()
    sys.exit(1)

CSV = CSV( Format_in, Format_out, Infile )
CSV.read()

CSV.write()