def get_rottentomatoes_ratings(title, year, result=None):
print "Processing {} - {}".format(title, year)
result = result or defaultdict(lambda: "N/A", {
'title': title,
'year': year
})
notes = []
try:
rt = RT(RT_API_KEY)
movie_list = rt.search(process_title(title))
if year:
movie_list = filter_year(movie_list, year)
if not movie_list:
raise Exception("No results found.")
try:
movie = movie_list[0]
result['rt_matched_title'] = movie['title']
result['rt_audience_score'] = movie['ratings']['audience_score']
result['rt_critics_score'] = movie['ratings']['critics_score']
except KeyError:
notes.append("Results not found: {}".format(title))
except Exception as e:
notes.append("Exception encountered: {}".format(e))
traceback.print_exc()
finally:
# result['Title'] = title
# result['Year'] = year
result['rt_notes'] = '|'.join(notes)
return result
def main():
assert os.environ["RT_KEY"], "Your Rotten Tomatoes API key should be stored in the RT_KEY env var!"
rt = RT() # NOTE: you should have your API key stored in RT_KEY before this will work
movies = []
link_template = ""
for country in BOX_OFFICE_COUNTRIES:
print "requesting box office hits for {}".format(country)
r = rt.lists('movies', 'box_office', limit=LIMIT, country=country)
movies += r['movies']
link_template = link_template or r['link_template']
time.sleep(10) # respect our API limits!
# to maintain compatibility with movies.json fields, our top level dict
# should have the following fields:
# total (int)
# movies (list)
# link_template (string)
total = len(movies)
result = {
"total": total,
"movies": movies,
"link_template": link_template,
}
with open(OUTPUT_FILE, "w") as f:
json.dump(result, f, indent=2, sort_keys=True)
def get_rottentomatoes_ratings(title, year, result=None):
print "Processing {} - {}".format(title, year)
result = result or defaultdict(lambda: "N/A", {'title':title, 'year': year})
notes = []
try:
rt = RT(RT_API_KEY)
movie_list = rt.search(process_title(title))
if year:
movie_list = filter_year(movie_list, year)
if not movie_list:
raise Exception("No results found.")
try:
movie = movie_list[0]
result['rt_matched_title'] = movie['title']
result['rt_audience_score'] = movie['ratings']['audience_score']
result['rt_critics_score'] = movie['ratings']['critics_score']
except KeyError:
notes.append("Results not found: {}".format(title))
except Exception as e:
notes.append("Exception encountered: {}".format(e))
traceback.print_exc()
finally:
# result['Title'] = title
# result['Year'] = year
result['rt_notes'] = '|'.join(notes)
return result
def __init__(self,
movie,
rotten_tomatoe_api_key,
tmdb_api_key,
aws_access_key,
aws_secret_key,
affiliate_key,
rt_id=None,
tmdb_id=None):
self._movie = movie
# amazon
self._amazon_product_search = AmazonProductSearch(
aws_access_key, aws_secret_key, affiliate_key)
# rotten tomatoes
self._rt = RT(rotten_tomatoe_api_key)
if rt_id:
self._rt_data = self._rt.info(rt_id)
else:
self._rt_data = self._rt.search(movie)[0]
# tmdb
self._tmdb = tmdb
self._tmdb.configure(tmdb_api_key)
movie = self._tmdb.Movies(
movie,
limit=True,
expected_release_date=self._rt_data['release_dates']
['theater']).get_best_match()
self._tmdb_data = self._tmdb.Movie(movie[1]['id'])
# youtube
self._yt_service = gdata.youtube.service.YouTubeService()
def test(request):
titles=[]
img_locations=[]
movie_locations=[]
extract_movies(titles, img_locations, movie_locations)
for i in range(len((titles))):
# print(titles[i])
# print(movie_locations[i])
# print(img_locations[i])
if (len(Movie.objects.filter(name=titles[i]))==0 and titles[i]!='LOL'):
if (len(RT(RT_KEY).search(titles[i]))>0 and RT(RT_KEY).search(titles[i])[0]['ratings']['critics_score']!=-1):
critics_score= RT(RT_KEY).search(titles[i])[0]['ratings']['critics_score']
audience_score=RT(RT_KEY).search(titles[i])[0]['ratings']['audience_score']
movie = Movie(name=titles[i], url=movie_locations[i]+'?country='+country, pic_url=img_locations[i], country=country, is_american=False, audience_score=audience_score, critics_score=critics_score)
movie.save()
# get_rotten_tomates()
else:
movie = Movie(name=titles[i], url=movie_locations[i]+'?country='+country, pic_url=img_locations[i], country=country, is_american=False)
movie.save()
# if (len(RT('bt7f4pcbku6m9mqzuhhncc9e').search(titles[i]))>0 and RT('bt7f4pcbku6m9mqzuhhncc9e').search(titles[i])[0]['ratings']['critics_score']!=-1):
# critics_score= RT('bt7f4pcbku6m9mqzuhhncc9e').search(titles[i])[0]['ratings']['critics_score']
# audience_score=RT('bt7f4pcbku6m9mqzuhhncc9e').search(titles[i])[0]['ratings']['audience_score']
# movie = Movie(name=titles[i], url=movie_locations[i]+'?country='+country, pic_url=img_locations[i], country=country, is_american=False, audience_score=audience_score, critics_score=critics_score)
# movie.save()
# else:
get_rotten_tomates()
t = get_template('index.html')
html = t.render(Context({}))
return HttpResponse(html)
def get_rotten_tomates(request):
for movie in Movie.objects.all():
print("got here")
if len(RT('bt7f4pcbku6m9mqzuhhncc9e').search(movie.name))==0:
break
movie_id=RT('bt7f4pcbku6m9mqzuhhncc9e').search(movie.name)[0]['id']
if (len(RT('bt7f4pcbku6m9mqzuhhncc9e').info(movie_id, 'reviews')['reviews'])>2):
reviewblob = RT('bt7f4pcbku6m9mqzuhhncc9e').info(movie_id, 'reviews')['reviews']
quote1=reviewblob[0]['quote']
fresh_bool1='fresh' in reviewblob[0]['freshness']
name1= reviewblob[0]['critic']
quote2=reviewblob[1]['quote']
fresh_bool2='fresh' in reviewblob[1]['freshness']
name2= reviewblob[1]['critic']
quote3=reviewblob[2]['quote']
fresh_bool3='fresh' in reviewblob[2]['freshness']
name3= reviewblob[2]['critic']
movie.quote1=quote1
print(movie.name)
print (movie.quote1)
movie.quote2=quote2
movie.quote3=quote3
movie.critic1=name1
movie.critic2=name2
movie.critic3=name3
movie.fresh1=fresh_bool1
movie.fresh2=fresh_bool2
movie.fresh3=fresh_bool3
movie.save()
t = get_template('index.html')
html = t.render(Context({}))
return HttpResponse(html)
def backend(movieTitles1,movieTitles2,database,svd_model_file): """ Returns recommended movies for two people movieTitles1: list (!) of movies person 1 wants to watch movieTitles2: list (!) of movies person 2 wants to watch """ #-----------------------------------------CREATE MOVIE ID AND TITLES DATAFRAME-----------------------------------------------------------# moviesDf=createMovieIDTitleDataFrame(database) #-----------------------------------------FIND MOVIEIDS OF INPUT MOVIES------------------------------------------------------------# # TO DO: Fix handling of not finding a movie ITEMIDS1=[getMovieID(moviesDf,movie) for movie in movieTitles1] ITEMIDS2=[getMovieID(moviesDf,movie) for movie in movieTitles2] #-------------------------------------------------GET SIMILARITIES-----------------------------------------------------------------# sims=getSimilarityMatrix(svd_model_file) #------------------------------------------------GET RECOMMENDED MOVIEIDS--------------------------------------------------------------# recsIDs = getRecMovieIDs(ITEMIDS1,ITEMIDS2,sims) #------------------------------------------------CONVERT TO TITLES--------------------------------------------------------------# recTitles=moviesDf.Title.loc[recsIDs].tolist() #------------------------------------------------GET POSTER URLS--------------------------------------------------------------# rt=RT() recMoviesInfo=[rt.search(title)[0] for title in recTitles] posterUrls=[movie['posters']['original'] for movie in recMoviesInfo] movieUrls=[movie['links']['alternate'] for movie in recMoviesInfo] #------------------------------------------------RETURN--------------------------------------------------------------# return zip(recTitles,posterUrls,movieUrls)
def sync_RT_movies(list = 'box_office'): # Lists could be: box_office, in_theaters, opening, and upcoming from rottentomatoes import RT rt = RT() movies = rt.movies(list) for movie in movies: movie_obj, created = Movie.objects.get_or_create(RT_movie_id = movie['id']) movie_obj.title = movie['title'] movie_obj.poster_detailed = movie['posters']['detailed'] movie_obj.save() print movie_obj print "********"
def dump_rt(self):
rttable = dict.fromkeys(KEYS_RTTABLE)
rt = RT("wy3s6eaj82m5ztwmsjuhnm38")
info = rt.feeling_lucky(search_term=self.name)
try:
rttable["rt_id"] = info["id"]
rttable["name"] = info["title"]
rttable["url"] = info["links"]["alternate"]
rttable["rating_audience"] = info["ratings"]["audience_score"]
rttable["rating_critics"] = info["ratings"]["critics_score"]
rttable["reviews_audience"] = info["ratings"]["audience_rating"]
rttable["reviews_critics"] = info["ratings"]["critics_rating"]
except Exception as e:
print "Error Fetching from RT : %s" % e
return rttable
def movie_rat(f):
movie=f[2]
if len(f)>3:
for i in range(3,len(f)):
movie+=' '+f[i]
rt = RT('53uhmdfpu5sybbb5y529skkh') #amitbj96
title=rt.search(movie)[0]['title']
D1=rt.search(title,page_limit=1)[0]['ratings']
D=rt.search(title)[0]['synopsis']
D=str(D);
if D=='':
D="No Synopsis Found."
# [movie name , ratings 5/5 , synopsis
LL= [title,str(D1['audience_score'])+"/100",str(D)]
return LL
class IMDBApi(object):
def __init__(self, title, year=0):
self.query = dict(title=title, year=int(year))
self.rt = RT(rotten)
def __call__(self):
matches = []
year = self.query['year']
results = self.rt.search(self.query['title'])
self.partial = copy.copy(results)
if year:
results = [x for x in results if x['year']==self.query['year']]
for r in results:
matches.append(self.rt.info(r['id']))
return matches
def _getAllInfo(self, search_params):
ret = []
try:
pretty_title = str(search_params)
source = RT(self.key)
results = source.search(pretty_title)
for i in results:
title = utils.sanitizeString(i.get("title", search_params.title))
year = str(i.get("year", search_params.year))
info = movie_types.MovieInfo(title, year)
#no genre without more effort
ret.append(info)
except Exception as ex: #bad need to find a better exception
utils.logWarning("Lib: {} Title: {} Error {}: {}".format(self.display_name, search_params.title, type(ex), ex),
title="{} lookup".format(self.display_name))
return ret
def rottenTomatoes(search_string, channel):
'''Takes a string and outputs details from rotten tomatoes'''
# Look up the json data from rotten tomatoes
movie = RT().search(search_string, page_limit=1)
# Respond if you don't find a match
if not movie:
ircMessage("no u", channel)
return
movie = movie[0]
ratings = movie['ratings']
# Not all movies have a synopsis
summary = ""
if movie['synopsis'] != "":
summary = "Summary: " + movie['synopsis'][:200] + "..."
# Movies without critics scores are replaced with N/A
critics = "N/A"
if ratings['critics_score'] > 0:
critics = ratings['critics_score']
message = "{0} ({1}): Critics: {2} Audience: {3} URL: http://rottentomatoes.com/m/{4} {5}".format(
movie['title'], movie['year'], critics, ratings['audience_score'],
movie['id'], summary)
ircMessage(message, channel)
def synopsis_fetcher(movies):
print movies
client = MongoClient()
db = client.recommender_db
movies_db = db.movies
synopsis = []
for movie in movies:
print movie['m_title']
synopse = {}
movie_in_db = movies_db.find_one({'m_id':movie['m_id']})
if not movie_in_db.has_key('sin'):
print movie['m_title']
s_rotten = RT().search(movie['m_title'])
sy_from_rotten = check_in_rotten(movie)
if sy_from_rotten:
synopse = sy_from_rotten
else:
search_imdb = search_title_imdb(movie['m_title'])
if search_imdb:
synopse = {'m_id': movie['m_id'], 'sin':search_imdb}
else:
synopsis.append(movie_in_db)
print "synopse append----------------------------"
if synopse:
synopsis.append(synopse)
print "----------------fetched synopse--------------"
print synopse
print "---------------------------------------------"
movies_db.update({"m_id":synopse['m_id']}, {'$set':{"sin":synopse['sin'] }})
return synopsis
class RTAdapter(Adapter):
"""Rotten Tomatoes Adapter
Implements the Rotten Tomatoes adapter.
"""
def __init__(self):
self.config = {
'api_key': '8yvmeqtydvquk9bxv4mvemhm',
}
self.rt = RT(self.config['api_key'])
def get_similar_film_titles(self, title):
# Get films
films = self.rt.search(title)[:5]
# Check if results are empty
if not films:
raise FilmNotFoundError()
# Extract titles
return [film['title'] for film in films]
def get_film(self, title):
# Get films
films = self.rt.search(title)
# Find film in recieved list
film_titles = [f.get('title', None) for f in films]
found_title = safe_find_film(title, film_titles)
# Raise error if not found
if not found_title:
raise FilmNotFoundError()
film = films[film_titles.index(found_title)]
return film
def get_film_score(self, title):
film = self.get_film(title)
# Check if ratings exists
if not 'ratings' in film:
return None
# Return film score
normalized_score = film['ratings']['critics_score'] / 100.0
return normalized_score
def __repr__(self):
return 'Rotten Tomatoes'
class RTAdapter(Adapter):
"""Rotten Tomatoes Adapter
Implements the Rotten Tomatoes adapter.
"""
def __init__(self):
self.config = {
'api_key': '8yvmeqtydvquk9bxv4mvemhm',
}
self.rt = RT(self.config['api_key'])
def get_similar_film_titles(self, title):
# Get films
films = self.rt.search(title)[:5]
# Check if results are empty
if not films:
raise FilmNotFoundError()
# Extract titles
return [film['title'] for film in films]
def get_film(self, title):
# Get films
films = self.rt.search(title)
# Find film in recieved list
film_titles = [f.get('title', None) for f in films]
found_title = safe_find_film(title, film_titles)
# Raise error if not found
if not found_title:
raise FilmNotFoundError()
film = films[film_titles.index(found_title)]
return film
def get_film_score(self, title):
film = self.get_film(title)
# Check if ratings exists
if not 'ratings' in film:
return None
# Return film score
normalized_score = film['ratings']['critics_score'] / 100.0
return normalized_score
def __repr__(self):
return 'Rotten Tomatoes'
def get_movie_rating(movie):
"""Returns a Rotten Tomatoes score for the given movie title"""
try:
json = RT.search(movie)[0]
title = json['title']
rating = json['ratings']['critics_score']
return (title, rating)
except IndexError:
return None
def sms():
response = twiml.Response()
body = request.form['Body']
rt = RT()
dump = rt.search(body)
if dump:
rating = dump[0]['ratings']['critics_score']
title = dump[0]['title']
msg = 'The film ' + str(title) + ' received a ' + \
str(rating) + ' on the TOMATOMETER.'
if rating > 50:
msg = msg + ' I would watch it.'
else:
msg = msg + ' Skip this one.'
else:
msg = "We didn't find " + body + " on RT... maybe check your spelling."
response.sms(msg)
return str(response)
def get_rotten_tomates():
for movie in Movie.objects.all():
if len(RT('bt7f4pcbku6m9mqzuhhncc9e').search(movie.name)) == 0:
break
movie_id = RT('bt7f4pcbku6m9mqzuhhncc9e').search(movie.name)[0]['id']
for j in range(
len(
RT('bt7f4pcbku6m9mqzuhhncc9e').info(
movie_id, 'reviews')['reviews'])):
reviewblob = RT('bt7f4pcbku6m9mqzuhhncc9e').info(
movie_id, 'reviews')['reviews']
quote = reviewblob[j]['quote']
fresh_bool = 'fresh' in reviewblob[j]['freshness']
name = reviewblob[j]['critic']
print(quote)
print(name)
review = Review(name=name,
body=quote,
fresh=fresh_bool,
movie=movie)
review.save()
def check_in_rotten(movie):
s_rotten = RT().search(movie['m_title'])
synopse = {}
if s_rotten:
m_rotten = s_rotten[0]
r_synopse = m_rotten['synopsis']
if r_synopse:
synopse = {'m_id': movie['m_id'], 'sin':r_synopse}
elif m_rotten.has_key('alternate_ids'):
i_id = m_rotten['alternate_ids']['imdb']
imdb_request_plot = get_imdb_plot(i_id)
if imdb_request_plot:
print imdb_request_plot
synopse = {'m_id': movie['m_id'], 'sin':imdb_request_plot}
else:
print "no movie plot or synopse "
return synopse
def __init__(self, movie, rotten_tomatoe_api_key, tmdb_api_key,
aws_access_key, aws_secret_key, affiliate_key, rt_id=None, tmdb_id=None):
self._movie = movie
# amazon
self._amazon_product_search = AmazonProductSearch(aws_access_key,
aws_secret_key, affiliate_key)
# rotten tomatoes
self._rt = RT(rotten_tomatoe_api_key)
if rt_id:
self._rt_data = self._rt.info(rt_id)
else:
self._rt_data = self._rt.search(movie)[0]
# tmdb
self._tmdb = tmdb
self._tmdb.configure(tmdb_api_key)
movie = self._tmdb.Movies(movie, limit=True,
expected_release_date=self._rt_data['release_dates']['theater']).get_best_match()
self._tmdb_data = self._tmdb.Movie(movie[1]['id'])
# youtube
self._yt_service = gdata.youtube.service.YouTubeService()
def test_empty_feeling_lucky_method_fails(self):
self.assertRaises(TypeError, RT().feeling_lucky)
def test_movies_box_office(self):
RT().movies('box_office')
path = call_args('path')
expected_path = '/api/public/v1.0/lists/movies/box_office.json'
self.assertEqual(path, expected_path)
def test_dvds_new_releases(self):
RT().dvds('new_releases')
path = call_args('path')
expected_path = '/api/public/v1.0/lists/dvds/new_releases.json'
self.assertEqual(path, expected_path)
def test_movies_in_theaters(self):
RT().movies('in_theaters')
path = call_args('path')
expected_path = '/api/public/v1.0/lists/movies/in_theaters.json'
self.assertEqual(path, expected_path)
def test_movies_opening(self):
RT().movies('opening')
path = call_args('path')
expected_path = '/api/public/v1.0/lists/movies/opening.json'
self.assertEqual(path, expected_path)
def test_initialized_api_key(self):
self.assertEqual(RT('called_api_key').api_key, 'called_api_key')
def __init__(self):
self.config = {
'api_key': '8yvmeqtydvquk9bxv4mvemhm',
}
self.rt = RT(self.config['api_key'])
def test_empty_search_url_keys(self):
RT().search('')
movie = call_args()
self.assertEqual(movie.keys(), ['apikey'])
def test_search_url_keys_with_page_arg(self):
RT().search('some movie', page=2)
movie = call_args()
self.assertEqual(movie.keys(), ['q', 'apikey', 'page'])
# -*- coding: utf-8 -*-
import time
from rottentomatoes import RT
rt = RT("uhaar3a93r8jqamjzmg5kum8")
import MySQLdb as mbd
# set up the connection to the MySQL database
con = mbd.connect('localhost', 'simmerin', 'simmerin', 'movies')
j = 0 # a counter for debugging purposes
with con:
cur = con.cursor()
# drop the table if it exists to prevent adding movies multiple times
cur.execute("DROP TABLE IF EXISTS rtid")
# create the table as detailed in the documentation
cur.execute("CREATE TABLE rtid(id INT PRIMARY KEY AUTO_INCREMENT, \
rtid INT)")
# bring in the titles and years from the Box Office Mojo data
cur.execute("SELECT * FROM mojo")
titles = cur.fetchall()
# loop through each title and pull out the rt id for the first film
# returned from the RT search with the correct year, if no matches
# use NULL for the rt id
for movie in titles:
print j
title = movie[1]
year = movie[5]
startTime = time.time()
searchResults = rt.search(title)
correctFilm = 0
def test_version_argument_with_string(self):
self.assertEqual(RT(version='2.5').version, '2.5')
#output the list of titles so I don't have to grab it again
#write_path = 'E:\Netflix Movies\\'
write_path = 'C:\Other Projects\Netflix Movies\\'
write_file = "titles.csv"
f = write_path+write_file
with open(f, 'w+') as my_file:
my_file.write('\n'.join(netflix_titles))
#Read back in the csv file:
netflix_titles = [line.rstrip('\n') for line in open(f)]
#set up RT api package
from rottentomatoes import RT
rt = RT('4cbst6rnnvresrd9e8q83hhs')
#just testing how RT's API works
fight_clubs = rt.search('101 dalmations')
for club in fight_clubs:
print "title="+club["title"]+" & ID="+club["id"] + " & released="+str(club["year"])
print "Critics' Score: "+str(club["ratings"]["critics_score"])
print "Audience Score: "+str(club["ratings"]["audience_score"])
if club["title"] in netflix_titles:
print "On Netflix!"
#scrape http://www.rottentomatoes.com/top/bestofrt/?year=2012 for best movies
from lxml import html
import requests
from rottentomatoes import RT
import os
import sys
import requests
import json
if len(sys.argv) != 3:
print "Usage: ./movie_rater.py dirname threshold"
exit(1)
searchdir = sys.argv[1]
keepthreshold = int(sys.argv[2])
myrt = RT()
movielist = os.listdir(searchdir)
movielist.sort()
def omdbsearch(title):
data = json.loads(requests.get("http://omdbapi.com/?t=" + title + "&r=json").text)
#print str(data)
return data
keepcount = 0
deletecount = 0
import config
import media
import my_movie_list
import youtube_search
# Rotten Tomatoes API for Python
# docs: https://github.com/zachwill/rottentomatoes
from rottentomatoes import RT
# initialize a rotten tomatoes instance
rt = RT( config.api_key )
# get a list of movie openings from RT
opening_movies = rt.lists('movies', 'opening')
# iterate through opening movies and create instances of our Movie class for each
movie_collection = []
for mov in opening_movies['movies']:
# Find youtube trailer
options = youtube_search.Options( mov['title'] + ' trailer', 1 )
trailer_id = youtube_search.search(options)
# Convert cast names to a comma separated list
cast = ''
l = mov['abridged_cast'].__len__() - 1
for i, actor in enumerate( mov['abridged_cast'] ):
cast += actor['name']
if i < l:
cast += ', '
movie = media.Movie(mov['title'], mov['synopsis'], mov['posters']['thumbnail'], trailer_id, mov['mpaa_rating'], mov['release_dates']['theater'], mov['ratings']['critics_score'], cast)
def test_first_json_loads_movies_result_is_returned(self):
data = RT().feeling_lucky('some movie')
self.assertEqual(data, 'first_result')
def test_search_url_keys_for_lion_king(self):
RT().search('the lion king')
movie = call_args()
assert 'my_api_key' in movie['apikey']
assert 'the lion king' in movie['q']
def test_uninitialized_api_key(self):
self.assertEqual(RT().api_key, 'my_api_key')
class MovieInfo(object):
def __init__(self, movie, rotten_tomatoe_api_key, tmdb_api_key,
aws_access_key, aws_secret_key, affiliate_key, rt_id=None, tmdb_id=None):
self._movie = movie
# amazon
self._amazon_product_search = AmazonProductSearch(aws_access_key,
aws_secret_key, affiliate_key)
# rotten tomatoes
self._rt = RT(rotten_tomatoe_api_key)
if rt_id:
self._rt_data = self._rt.info(rt_id)
else:
self._rt_data = self._rt.search(movie)[0]
# tmdb
self._tmdb = tmdb
self._tmdb.configure(tmdb_api_key)
movie = self._tmdb.Movies(movie, limit=True,
expected_release_date=self._rt_data['release_dates']['theater']).get_best_match()
self._tmdb_data = self._tmdb.Movie(movie[1]['id'])
# youtube
self._yt_service = gdata.youtube.service.YouTubeService()
def get_amazon_purchase_links(self, top_cast, runtime):
products = self._amazon_product_search.item_search(self._movie,
top_cast, runtime)
return products
@property
def cast(self):
'''
Returns the names of the full cast for this movie
'''
full_cast = self._rt.info(self._rt_data['id'], 'cast')
names = [cast['name'] for cast in full_cast['cast']]
return names
@property
def critic_reviews(self):
'''
Returns a list of critic reviews for this movie. The list
is componsed of json document.
'''
reviews = self._rt.info(self._rt_data['id'], 'reviews')
return reviews['reviews']
@property
def critics_score(self):
'''
Returns the rotten tomatoe critic score for this movie
'''
return self._rt_data['ratings']['critics_score']
@property
def director(self):
'''
Returns a list of directors for this movie
'''
return self._tmdb_data.get_director()
@property
def genres(self):
'''
Returns the genres of this movie, supplied by tmdb
'''
genres = self._tmdb_data.get_genres()
genres = [genre['name'].lower() for genre in genres]
return genres
@property
def imdb_id(self):
'''
Returns a list of directors for this movie
'''
try:
return "tt" + self._rt_data['alternate_ids']['imdb']
except:
return self._tmdb_data.get_imdb_id()
@property
def poster(self):
'''
Returns the poster of the movie, in its original size
'''
return self._tmdb_data.get_poster()
@property
def runtime(self):
'''
Return the runtime of this movie in minues
'''
try:
return int(self._rt_data['runtime'])
except:
return int(self._tmdb_data.get_runtime())
@property
def release_date(self):
'''
Returns this movie's release date in {year}-{month}-{day} format
'''
try:
return parser.parse(self._rt_data['release_dates']['theater'])
except:
return parser.parse(self._tmdb_data.get_release_date())
@property
def similar_movies(self):
'''
Returns a list of imdb ids of movies that are similar to this one
'''
movies = self._rt.info(self._rt_data['id'], 'similar')['movies']
# if movie is none or len(movies) == 0:
# search google current movie title +
# parse DOM for
return movies
@property
def synopsis(self):
'''
Returns this movie's synopsis
'''
synopsis = self._rt_data['synopsis']
if len(synopsis) == 0:
synopsis = self._tmdb_data.get_overview()
return synopsis
@property
def title(self):
'''
Returns this movie's title
'''
return self._rt_data['title']
@property
def trailers(self, limit=3):
'''
This function returns a list of trailers for this movie.
We will use TMDB's data if it returns 3 or more trailers. If not,
we will query youtube with the search term: "{movie_name} trailer
{release_year}" to find trailers for this movie.
Returns a list of youtube ids of the trailers
'''
trailers = self._tmdb_data.get_trailers()['youtube']
if len(trailers) > limit:
return [trailer['source'] for trailer in trailers]
else:
release_year = str(self.release_date).split('-')[0]
query = gdata.youtube.service.YouTubeVideoQuery()
query.vq = "{title} trailer {release_year} ".format(
title=self._movie, release_year=release_year)
query.orderby = 'relevance'
feed = self._yt_service.YouTubeQuery(query)
entries = self._remove_long_youtube_videos(feed.entry[:3])
entries = self._remove_unrelated_videos(entries)
unique_entries = self._remove_duplicate_youtube_videos(entries)
return unique_entries
def _remove_duplicate_youtube_videos(self, entries, threshold=5):
'''
This method removes duplicate videos by measuring the runtime
of the youtube videos.
If two videos are within 5 seconds (the threshold) of each
other in runtime, we assume that one of the videos is a duplicate
of the other.
'''
limit = 3
videos = []
for entry in entries:
runtime = int(entry.media.duration.seconds)
similar = [runtime >= int(video["runtime"]) - threshold and
runtime <= int(video["runtime"]) + threshold for
video in videos]
if not any(similar):
video_id = self._extract_youtube_id(entry.media.player.url)
videos.append({"yt_id": video_id, "runtime": runtime})
yt_ids = [video['yt_id'] for video in videos]
return yt_ids
def _remove_long_youtube_videos(self, entries, max_seconds=600):
entries = filter(lambda entry:
int(entry.media.duration.seconds) < max_seconds,
entries)
return entries
def _remove_unrelated_videos(self, entries):
entries = filter(lambda entry:
fuzzywuzzy.fuzz.ratio(entry.media.title.text.decode('utf-8').lower(), self._movie.lower()) > 20,
entries)
return entries
def _extract_youtube_id(self, youtube_url):
video_id = youtube_url.split('v=')[1]
ampersand_position = video_id.find('&')
if(ampersand_position != -1):
video_id = video_id[0:ampersand_position]
return video_id
def test_version_argument_with_float(self):
self.assertEqual(RT(version=2.5).version, '2.5')
import mechanize
import settings
import time
from bs4 import BeautifulSoup
from Movie import db, Movie
from movie_crawler import save_movie_info_to_mongo
from rottentomatoes import RT
if __name__ == "__main__":
db_conn_settings = dict([(k.lower(), v) for k, v in
settings.ProdConfig.MONGODB_SETTINGS.items() if v])
db.connect(**db_conn_settings)
rotten_tomatoe_api_key = settings.Config.ROTTEN_TOMATOES_API_KEY
rt = RT(rotten_tomatoe_api_key)
movies = Movie.objects.timeout(False)
for index, movie in enumerate(movies):
print index
if len(movie.similar_movies) == 0:
print "updating: %s %s" % (movie.title, movie.release_date)
# creating our search term
search_term = movie.title.replace(" ", "+")
url = "https://www.google.com/search?q=%s+Movie" % search_term
# querying google
br = mechanize.Browser()
br.addheaders = [('user-agent', 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/30.0.1599.69 Safari/537.36'),
('accept', 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8')]
br.set_handle_robots(False)
def test_nonempty_search_url_path(self):
RT().search('some movie')
path = call_args('path')
self.assertEqual(path, '/api/public/v1.0/movies')
from rottentomatoes import RT
rt = RT()
movies = rt.movies('in_theaters')
import pdb; pdb.set_trace()
def test_nonempty_search_url_keys(self):
RT().search('some movie')
movie = call_args()
self.assertEqual(movie.keys(), ['q', 'apikey'])
def test_search_results_for_standard_datatype(self):
results = RT().search('some movie')
self.assertEqual(results, ['first_result', 'second_result'])
def test_search_url_keys_with_multiple_kwargs(self):
RT().search('some movie', page=2, page_limit=5)
movie = call_args()
self.assertEqual(movie.keys(), ['q', 'apikey', 'page', 'page_limit'])
#!/usr/bin/python
from rottentomatoes import RT
import sys
import pprint
#api_key = 'bpbjuznunrqvkeuqjk9m2dmm'
#RT(api_key).search('gone with the wind')
rt = RT()
dump = rt.search(sys.argv[1])
pprint.pprint(dump)
if dump:
rating = dump[0]['ratings']['critics_score']
title = dump[0]['title']
msg = 'The film ' + str(title) + ' received a ' + \
str(rating) + ' on the TOMATOMETER.'
if rating > 50:
msg = msg + ' I would watch it.'
else:
msg = msg + ' Skip this one.'
else:
msg = 'nothing here!'
print msg
def test_search_url_keys_for_ronin(self):
RT().search('ronin')
movie = call_args()
assert 'my_api_key' in movie['apikey']
assert 'ronin' in movie['q']
from rottentomatoes import RT
import time
rt = RT()
# movies = ['fight club','gravity','toy story 3','american hustle','skyfall','jack and jill','basic instinct 2','white out','lost souls','babylon','argo','bears','her','up','a beautiful mind','braveheart','the hurt locker','gambit','paranoia','getaway']
movies = [line.strip("\n") for line in open("movies.txt")]
f = open("reviews-date.txt", "w")
count = 0
for u in movies:
movlst = rt.search(u, page_limit=1)
if movlst:
if movlst[0][u"id"] != "":
review = rt.info(movlst[0][u"id"], "reviews")
review.viewkeys()
rlt = review[u"reviews"] # review is a dict, value of each key is list, each list element is a dict
for a in rlt:
f.write(a[u"quote"] + "\t" + a[u"date"] + "\n")
f2.write(a[u"date"] + "\n")
count = count + 1
# print a[u'quote']
if len(rlt) == 0:
print u
else:
print count
else:
print u
else:
print u
time.sleep(5)
f.close()
#!/usr/bin/env python
from rottentomatoes import RT
import json, csv
rt = RT()
#movies = ['toy story 3', 'the lion king', 'the matrix', 'the dark knight', 'inception', 'titanic', 'the godfather', 'the little mermaid', 'the shining', 'avatar']
movies = ['pocahontas', 'shawshank redemption']
json = []
for movie in movies:
json = json + rt.search(movie)
with open('test2.csv', 'wb+') as f:
dict_writer = csv.DictWriter(
f,
fieldnames=[
'ratings', 'abridged_directors', 'links', 'title',
'critics_consensus', 'release_dates', 'abridged_cast', 'synopsis',
'mpaa_rating', 'year', 'alternate_ids', 'posters', 'runtime', 'id'
])
dict_writer.writeheader()
dict_writer.writerows(json)
def __init__(self, title, year=0):
self.query = dict(title=title, year=int(year))
self.rt = RT(rotten)
