actualize-data.py

#!/usr/bin/python

# This script processes a JSON data file and updates all dates in the demo data.
# It does this by comparing the dates in the demo data with a reference date for the data set.
# All dates are updated to be exactly the same number of days from today as they were from the reference date (either in the future or in the past).
# This updates the demo data to always be "current".

import json
import sys
import re
import datetime 
from datetime import date

if (len(sys.argv) < 3):
  print "Usage: actualize-data.py [datafile] [output file]"
  exit(1)

dataFile = sys.argv[1]
with open(dataFile) as f:
    data = json.load(f)
    dataSetReferenceDate = datetime.datetime.strptime(data[0]['dueDate'].split('T')[0], "%Y-%m-%d").date()
outputFile = sys.argv[2]

inFile = open(dataFile)
outFile = open(outputFile, 'w+')
p = re.compile("([0-9]{4,4}-[0-9]{2,2}-[0-9]{2,2})T[0-9]{2,2}:[0-9]{2,2}:[0-9]{2,2}")

currentDate = date.today()
deltaDays = currentDate - dataSetReferenceDate

print "Updating XL release data set.\n"
print "Data set filename: " + str(dataFile)
print "Data set reference date: " + str(dataSetReferenceDate)
print "Current date: " + str(currentDate)
print "Delta to apply: " + str(deltaDays)
print "Output filename: " + str(outputFile)

for line in inFile:
  line = line.rstrip()
  m = p.search(line)
  if (m):
    dataSetDate =  datetime.datetime.strptime(m.group(1), "%Y-%m-%d").date()
    #print "Ref date: " + str(dataSetReferenceDate) + ", days since ref date: " + str(deltaDays)
    newDataSetDate = dataSetDate + deltaDays
    #print "Data set date: " + str(dataSetDate) + ", new date: " + str(newDataSetDate)
    print >>outFile, re.sub(m.group(1), str(newDataSetDate), line)
  else:
    print >>outFile, line

inFile.close()
outFile.close()

print "Done."