parsexml.py

#!/usr/bin/env python3
import csv
import xml.etree.ElementTree as ET
import sys
import os.path


class QuestionData:
    final_answer = ""
    final_answer_time = 0
    first_answer = ""
    attempts = 0
    first_answer_time = 0

    def __init__(self,final_answer,final_answer_time,attempts,first_answer,first_answer_time):
        self.final_answer = final_answer
        self.final_answer_time = final_answer_time
        self.first_answer = first_answer
        self.first_answer_time = first_answer_time
        self.attempts = attempts


def xml_to_csv(xml_filename):
    """
    Parses .xml files generated by newer versions of iClicker software in SessionData
    A CSV file will be written to the same path as the XML file, so it is important that any path, be it
    absolute or relative, is included in the xml_filename argument. The CSV file is not a perfect replica of
    older (i.e. iClicker 6) CSV files, but is our best approximation at this time. It should be enough for
    Rainbow Grades to function properly.
    """

    csv_filename = xml_filename[:-3] + "csv"
    try:
        with open(xml_filename,"r") as readfile:
            tree = ET.parse(xml_filename)
            root = tree.getroot()

            questions_in_order = []
            start_times = {}
            stop_times = {}
            user_question_data = {}

            for child in root:
                if child.tag == "p":  # This is a polling tag
                    question = child.attrib["qn"]
                    start_times[question] = child.attrib["strt"]
                    stop_times[question] = child.attrib["stp"]
                    questions_in_order.append(question)
                    question_votes = {}
                    for qchild in child:
                        if qchild.tag == "v":  # This is a voting tag
                            clicker_id = qchild.attrib["id"]
                            if clicker_id not in user_question_data:
                                user_question_data[clicker_id] = {}
                            user_question_data[clicker_id][question] = {}
                            if "fans" in qchild.attrib:
                                user_question_data[clicker_id][question] = QuestionData(qchild.attrib["ans"],
                                                                                     qchild.attrib["fanst"],
                                                                                     qchild.attrib["att"],
                                                                                     qchild.attrib["fans"],
                                                                                     qchild.attrib["tm"])

                            question_votes[clicker_id] = qchild.attrib["ans"]

            with open(csv_filename, 'w') as writefile:
                csvwriter = csv.writer(writefile)  # Need to change dialect to be iclicker compliant

                # Write the header
                # Right now we don't have min reply/min correct in XML land, instead we have MinPart_S
                next_row = ["Scoring"]
                if "perf" in root.attrib:
                    performance = root.attrib["perf"]
                else:
                    performance = -1

                if "part" in root.attrib:
                    participation = root.attrib["part"]
                else:
                    participation = 1
                csvwriter.writerow(["Scoring", "Performance = " + performance,
                                    "Participation = " + participation, "Min Reply = 2",
                                    "Min Correct = 0",
                                    " "])
                next_row = ["Question", " ", " "]
                for i in range(len(questions_in_order)):
                    next_row = next_row + ["Question " + str(i + 1), "Score", "Final Answer Time", "Number of Attempts",
                                           "First Response", "Time"]
                csvwriter.writerow(next_row)
                next_row = ["Start Time", " ", " "]
                for question in questions_in_order:
                    next_row = next_row + [" " + start_times[question], " ", " ", " ", " ", " "]
                csvwriter.writerow(next_row)

                next_row = ["Stop Time", " ", " "]
                first_stop = True
                for question in questions_in_order:
                    if not first_stop:
                        next_row = next_row + [" " + stop_times[question], " ", " ", " ", " ", " "]
                    else:
                        next_row = next_row + [stop_times[question], " ", " ", " ", " ", " "]
                        first_stop = False
                csvwriter.writerow(next_row)

                next_row = ["Correct Answer", " ", " "]
                first_stop = True
                for question in questions_in_order:
                    if not first_stop:
                        next_row = next_row + [" ", " ", " ", " ", " ", " "]
                    else:
                        next_row = next_row + ["", " ", " ", " ", " ", " "]
                        first_stop = False
                csvwriter.writerow(next_row)

                for user in sorted(user_question_data.keys()):
                    next_row = [user, "", "0"]
                    for question in questions_in_order:
                        if question in user_question_data[user]:
                            qd = user_question_data[user][question]
                            next_row = next_row + [qd.final_answer, 0, qd.final_answer_time, qd.attempts,
                                                   qd.first_answer, qd.first_answer_time]
                        else:
                            next_row = next_row + ["", "", "", "", "", ""]
                    csvwriter.writerow(next_row)

    except IOError as e:
        print("File I/O error: {}".format(e))
        exit(-1)


if __name__ == '__main__':
    if len(sys.argv) != 2:
        print("Correct usage is {} [file with iclicker {\"file\":...} entries]".format(sys.argv[0]))
        exit(-1)

    files = []

    try:
        with open(sys.argv[1]) as json_file:
            for line in json_file:
                # Extract just the filenames of the session data
                files += [x.strip()[1:-1] for x in line.split("[")[1].split("]")[0].split(",")]
    except IOError as e:
        print("Error reading JSON excerpt: {}".format(e))

    for filename in files:
        if len(filename) >= 4 and filename[-4:] == ".xml":
            xml_to_csv(filename)