psych.py

from __future__ import division, print_function
import csv

def get_weight(text_importance):
    '''returns the integer value of the
    text based importance rating'''

    imp = {"Irrelevant":0,
           "A Little Important":1,
           "Somewhat Important":10,
           "Very Important":50,
           "Mandatory":250}
    return imp[text_importance]

def load_data(filename):
    with open(filename) as csvfile:
        people = []
        spamreader = csv.reader(csvfile, delimiter=',')
        row_num = 0
        for row in spamreader:
            responses = []
            if row[0] == '':
                continue
            if row_num>0:
                name = row[1]
                gender = row[2]
                for i in range(3,len(row),3):
                    answer = row[i]
                    preferred = row[i+1].split(", ")
                    weight = get_weight(row[i+2])

                    responses.append(Response(answer,
                                              preferred,
                                              weight,
                                              i))
                people.append(Student(name,
                                      gender,
                                      responses))
            
            row_num+=1

        return people
class Response:
    def __init__(self,
                 personal_answer,
                 acceptable_answers,
                 weight,
                 question_id):
        self.personal_answer = personal_answer
        self.acceptable_answers = acceptable_answers
        self.weight = weight
        self.question_id = question_id

def get_match_percentages(person_1,
                         person_2):
    person_1_score = 0
    total_possible_points_person_1 = person_2.total_points

    person_2_score = 0
    total_possible_points_person_2 = person_1.total_points

    #makes math hard if we go equal rights, sorry
    assert (person_1.gender != person_2.gender)

    #time complexity of algorithm O(n^2)
    #time saved in thought O(n^2)
    for person_1_response in person_1.responses:
        for person_2_response in person_2.responses:
            if(person_1_response.question_id == person_2_response.question_id):
                
                if person_1_response.personal_answer in person_2_response.acceptable_answers:
                    person_1_score += person_2_response.weight
                
                if person_2_response.personal_answer in person_1_response.acceptable_answers:
                    person_2_score += person_1_response.weight
    
    person_1_percentage = person_1_score / total_possible_points_person_1

    person_2_percentage = person_2_score / total_possible_points_person_2

    
    assert len(person_2.responses) == len(person_1.responses)

    root = 1/len(person_2.responses)
    return (person_1_percentage * person_2_percentage) ** (1/2)
    return (person_1_percentage * person_2_percentage) ** (root)

class Student:
    def __init__(self,name,gender,responses):
        self.name = name
        self.gender = gender
        self.responses = responses
        self.total_points = 0
        for response in self.responses:
            self.total_points += response.weight
    def __str__(self):
        return str(self.name)
    def __unicode__(self):
        return self.name
def sep_genders(data):
    males = []
    females = []
    for person in data:
        if person.gender == "Male":
            males.append(person)
        else:
            females.append(person)
    return [males,females]
def find_all_matches(gender_sep_data):
    males = gender_sep_data[0]
    females = gender_sep_data[1]

    for person in males:
        for other_person in females:
            percent = get_match_percentages(person,
                                            other_person)
            print("{0} and {1} : {2}%".format(person.name,
                                             other_person.name,
                                              percent*100))
        print()

    print()

    for person in females:
        for other_person in males:
            percent = get_match_percentages(person,
                                            other_person)

            print("{0} and {1} : {2}%".format(person.name,
                                              other_person.name,
                                              percent*100))
        print()

data = load_data("psych.csv")
find_all_matches(sep_genders(data))