gistfile1.txt

"""
koda: b2
skupina: b
calibration_num_letters: =calibration8 level, max_num_letters?
time_calibration_num_letters: average time of elapsed time
choice: int(bool(["results"]["choice"] == "hard"))
correct: ["results"]["expsolved"]
time: ["results"]["elapsedTime"]
"""

import glob
import json
import re
from decimal import Decimal
from datetime import datetime


def parse_data(data):
    elapsed_time = Decimal(0)
    for result in data["calibration8"]["results"]:
        elapsed_time = elapsed_time + Decimal(result["elapsedTime"])

    output_data = {}
    output_data["time_cal_num_let"] = Decimal(elapsed_time)/len(data["calibration8"]["results"])

    for level in data.keys():
        if "letters" in level:
            output_data["{}{}".format(level, "choice")] = int(bool(data[level]["choice"] == "hard"))
            output_data["{}{}".format(level, "correct")] = int(bool(data[level]["results"][0]["expsolved"] == True))
            output_data["{}{}".format(level, "time")] = data[level]["results"][0]["elapsedTime"]

    return output_data

experiment_data = []
user_data = {}

for filename in glob.glob("*.json"):
    re_txt = re.search("(.*?)\.json", filename)
    if re_txt:
        user_id = re_txt.group(1)
        user_data[user_id] = json.loads(open(filename).read())

        print "parsing user id {}".format(user_id)
        try:
            choices = {}
            choices = parse_data(user_data[user_id][0]["levels"])
            choices["uid"] = user_id
            choices["cal_num_let"] = user_data[user_id][0]["max_num_letters"]

            if not choices.get("date_of_experiment"):
                choices["date_of_experiment"] = datetime.fromtimestamp(user_data[user_id][0].get("levels", {}).get("time"))

            experiment_data.append(choices)
            print choices
        except Exception as e:
            print e.message
            print "!!!---- Could not parse user id {}----!!!".format(user_id)
            pass



import csv

{
  'letters12choice': 1,
  'letters12time': 3311,
  'letters19time': 2704,
  'letters12correct': 0,
  'letters14correct': 1,
  'letters21choice': 0,
  'uid': 'b97',
  'letters23choice': 0,
  'letters21correct': 1,
  'letters23correct': 1,
  'letters19choice': 1,
  'letters16correct': 1,
  'time_cal_num_let': Decimal('2158.388888888888888888888889'),
  'letters23time': 2246,
  'letters16time': 1561,
  'letters14choice': 0,
  'cal_num_let': 7,
  'letters16choice': 0,
  'letters14time': 3654,
  'letters21time': 1523,
  'letters19correct': 1
}

number_mappings = {
    "12": "1_1",
    "14": "1_2",
    "16": "2_1",
    "19": "2_2",
    "21": "3_1",
    "23": "3_2" 
}


with open("rezultati.csv", 'wb') as myfile:
    wr = csv.writer(myfile, delimiter='\t')
    headers_str = "Koda,skupina,cal_num_let,time_cal_num_let,choice1_2,correct1_2,time1_2,choice2_2,Correct2_2,time2_2,choice3_2,Correct3_2,time3_2,Choice1_1,Correct1_1,time1_1,Choice2_1,Correct2_1,time2_1,Choice3_1,Correct3_1,time3_1,date_of_experiment".lower()
    headers = headers_str.split(",")

    empty = range(len(headers))
    empty = map(lambda a: "XXXX", empty)

    for record in experiment_data:
        for key in record.keys():
            if "choice" in key:
                number = key[7] + key[8]
                h = "choice" + number_mappings[number]
                empty[headers.index(h)] = record[key]
            if "correct" in key:
                number = key[7] + key[8]
                h = "correct" + number_mappings[number]
                empty[headers.index(h)] = record[key]
            if "time" in key and "time_cal_num_let" not in key:
                number = key[7] + key[8]
                h = "time" + number_mappings[number]
                empty[headers.index(h)] = record[key]
            if "uid" in key:
                empty[headers.index("koda")] = record[key][0]
                empty[headers.index("skupina")] = record[key]
            if "cal_num_let" in key:
                empty[headers.index("cal_num_let")] = record[key]
            if "time_cal_num_let" in key:
                empty[headers.index("time_cal_num_let")] = int(float(record[key]))
            if "date_of_experiment" in key:
                empty[headers.index("date_of_experiment")] = record[key]

        if "XXXX" in empty:
            print empty
            raise Exception
        wr.writerow(empty)