Added IBJoinHistoricalSeries.py

IBCombineHistoricalData.py

@@ -27,8 +27,8 @@ def parseLine(line):
     return dateStr.strip(), priceStr.strip(), closePriceStr.strip()
 
 if __name__ == "__main__":
-    dataPath = "C:/Dropbox/CninSrc/JTS/TWS API/samples/Java/Data/CAD/BID/"
-    outFile = "C:/Dropbox/CninSrc/JTS/TWS API/samples/Java/Data/CAD_BID.txt"
+    dataPath = "C:/Dropbox/CninSrc/JTS/TWS API/samples/Java/Data/CAD/ASK/"
+    outFile = "C:/Dropbox/CninSrc/JTS/TWS API/samples/Java/Data/CAD_ASK.txt"
 
     files = os.listdir(dataPath)
     files.sort(key=tokenize) #this allows files to be sorted in natural order eg 1.txt, 2.txt ... 10.txt instead of 1.txt 10.txt 2.txt etc
@@ -43,6 +43,8 @@ def parseLine(line):
         lines = f.readlines()
         for line in lines:
             date, time, closePrice = parseLine(line)
-            out.write(date + ", " + time + ", " + closePrice + "\n")
+            time = time.replace(":", "")
+            datetime = str(long(date + time))
+            out.write(datetime + ", " + closePrice + "\n")
 
     out.close()

IBDownloadHistoricalData.java

@@ -54,11 +54,11 @@ class IBDownloadHistoricalData  implements EWrapper {
     private FileLog     mServerErrorsLog = new FileLog("ServerErrors.txt");
 
 	//CHANGE this to the desired path
-	private String 		mDataPath = "Data/CAD/BID/";
+	private String 		mDataPath = "Data/JPY/BID/";
 	//CHANGE this to the desired contract
 	private Contract mContract =  new Contract(0, "USD", "CASH", "",
                     0, "", "",
-                    "IDEALPRO", "CAD", "", "",
+                    "IDEALPRO", "JPY", "", "",
                     new Vector<ComboLeg>(), "IDEALPRO", false,
                     "", "");
 	//CHANGE this between BID and ASK to get different fields

IBJoinHistoricalSeries.py

@@ -0,0 +1,28 @@
+import numpy as np
+import numpy.lib.recfunctions as rfn
+import pandas as pd
+
+#The logic in this file takes many the data files generated by IBCombineHistoricalData
+#"joins" them by datetime to create a file with one date-time column and numerous data columns
+if __name__ == "__main__":
+    #MANUALLY SET THESE FILES FOR JOINING
+    filesToJoin = [
+        "C:/Dropbox/CninSrc/JTS/TWS API/samples/Java/Data/CAD_BID.txt",
+        "C:/Dropbox/CninSrc/JTS/TWS API/samples/Java/Data/CAD_ASK.txt"
+    ]
+    #MANUALLY SET THIS FOR OUTPUT FILE
+    outFile = "C:/Dropbox/CninSrc/JTS/TWS API/samples/Java/Data/JOINED.txt"
+    #MANUALLY SET THIS HEADER SO EASIER TO REMEMBER COLUMNS
+    headerTxt = "Time,CAD_BID,CAD_ASK"
+
+    joined = []
+    for file in filesToJoin:
+        print "Handing: " + file
+        data = np.genfromtxt(file, delimiter = ',', dtype=np.dtype([('time',np.long), ('price', np.float)]))
+        print "Joining"
+        if len(joined) == 0:
+            joined = data
+        else:
+            joined = rfn.join_by('time', joined, data, jointype='inner', usemask=False)
+
+    np.savetxt(outFile, joined, delimiter=',', fmt="%s", header=headerTxt, comments="")