Data Science: What is the near future of Superheroines?

gender_analysis_on_movies.py

import collections
import re


def get_first_name(aString):
    if not aString:
        return aString
    ss = aString.replace('*', '').split(' ')  # ['Leonard', 'Nimoy*Chris', 'PineZachary', 'QuintoZoe', 'SaldanaKarl']
    name_list = [] # result returned for this function
    for name in ss:
        names = re.findall('([A-Z])', name) #[N, C]
        if len(names) == 1: # L
            name_list.append(name)
        elif len(names) == 2:
            name_list.append(name[name.find(names[1],1):])  #name.find(names['C']) = 5 --> name[5:] 
    return name_list[0:-1] # array 

def read_marvel(file_name):
    movies = []
    with open(file_name) as f:
        movies = f.read().split('\n')
    return movies

def read_csv(file_name): # file_name = 'boxoffice.csv'
    movie_list = [] # create an empty list
    with open(file_name) as f:
        txt = f.read()
        row_list = txt.split('\n') # a list of each row of data
        for row in row_list:
            name_list = row.split(',')
            if len(name_list) >= 3:
                name_list[1] = get_first_name(name_list[1])
                # 0: movie name, 1: actor names, 2: year
                movie_list.append((name_list[0], name_list[1], name_list[2]))
    return movie_list

def lookup_gender(filename):
    gender_dict = {}
    with open(filename) as f:
        firstname_gender = f.read().split('\n')
        for t in firstname_gender[:-1]: # there is a '' at the end cuz rows were split by \n
            firstname, gender = t.split(',')
            gender_dict[firstname] = gender
    return gender_dict

## SCRIPT begins
# construct a first name to gender dictionary
gender_dict = lookup_gender('name.csv')

movie_list = read_csv('boxoffice.csv')  # movie_list has 3 columns: name, actor list, year
all_movie_dict = collections.OrderedDict()
sorted_movie_list = sorted(movie_list, key=lambda x:x[2], reverse=True)

##all_actors = set()
##for m in sorted_movie_list:
##    for n in m[1]:
##        all_actors.add(n)
##
##with open('raw_name.csv','w') as f:
##    for name in all_actors:
##        f.write(name + '\n')


## Analysze all movie's actor gender by year
all_year_dict = collections.OrderedDict()
for m in sorted_movie_list:
    all_movie_dict[m[0]] = [m[1], m[2]]
    year = m[2]
    actors = m[1]
    try:
        genders = [gender_dict[name] for name in actors]
    except KeyError:
        pass
    if year in all_year_dict:
        all_year_dict[year] += genders
    else:
        all_year_dict[year] = genders

print ' key: year, value: list of gender of male/female'
for i in all_year_dict:
    my_temp_dict = {j:all_year_dict[i].count(j) for j in all_year_dict[i]}
    print i, my_temp_dict


## Analyze Marvel's movie actor gender by year
# construct a dict, key: year, value: list of gender of male/female
year_dict = collections.OrderedDict()
for marvel_movie in read_marvel('marvel_movies.txt'): # m as movie name
    # for each name in the marvel movie
    year = str(all_movie_dict[marvel_movie][1])
    actors = all_movie_dict[marvel_movie][0]
    gender = [gender_dict[name] for name in actors]
    if year in year_dict:
        year_dict[year] += gender
    else:
        year_dict[year] = gender

# construct a dict, key: male/female, value: count
print 'key: male/female, value: count'
for i in year_dict:
    my_dict = {j:year_dict[i].count(j) for j in year_dict[i]}
    print i, my_dict