class Watcher:
def __init__(self):
self.imdb = Imdb(anonymize=True)
self.tracked_shows = self.get_shows()
self.static_dir = os.path.join(os.path.dirname(__file__),
'../static/images')
def get_shows(self):
"""
gets all current popular shows from imdb
"""
shows = self.imdb.popular_shows()
tracked_shows = []
for show in shows:
tracked_shows_d = {}
tracked_shows_d['id'] = show['tconst']
tracked_shows_d['title'] = show['title']
tracked_shows_d['poster'] = show['image']['url']
tracked_shows.append(tracked_shows_d)
return tracked_shows
def get_show_id(self, show_title):
"""
Gets show title id
args:
show_title: name of show to be queried
returns:
show_id: id of show
"""
for show in self.tracked_shows:
if show_title == show['title']:
return show['id']
def get_episodes(self, show_id):
"""
Gets all episodes of a given show
args:
show_id: tconst id from imdb
returns:
ist of episodes
"""
return self.imdb.get_episodes(show_id)
def get_all_episodes(self):
"""
Gets all episodes
args:
None
returns:
list of episodes for all shows"""
programs = {}
for show in self.tracked_shows:
programs[show['title']] = self.get_episodes(show['id'])
return programs
def get_poster(self, show_title):
"""
gets the img url for the poster of a show
args:
show_title: title of show
returns:
dictionary with {show_title: poster_url}
"""
#print(self.tracked_shows[show_id]['poster'])
for show in self.tracked_shows:
if show['title'] == show_title:
return {show_title: show['poster']}
def save_posters(self, urls, title):
title = self.sanitize_title(title)
dest = '{}/{}.jpg'.format(self.static_dir, title)
urllib.request.urlretrieve(url, dest)
def sanitize_title(self, title):
forbidden = ('<', '>', ':', '"', '/', '\\', '|', '?', '*')
for char in forbidden:
title = title.replace(char, '')
return title
def get_show_titles(self):
"""
Gets show titles
args:
None
returns:
list of show titles
"""
return [show['title'] for show in self.tracked_shows]
class IMDBGraph:
def __init__(self, anonymize=True):
self._imdb = Imdb(anonymize=anonymize, cache=True)
self._graph = nx.Graph()
def _add_node(self, name, nodetype):
''' Add simple node without attributes
'''
if name not in self._graph.nodes():
self._graph.add_node(name, node=nodetype)
def addPerson(self, idname):
''' add New actor/actress no the graph
'''
actor = self._imdb.get_person_by_id(idname)
self._graph.add_node(actor.name)
def addMovie(self, idname):
movie = self._imdb.get_title_by_id(idname)
self._add_node(movie.title, 'movie')
logging.info("Loading {0}".format(idname))
def addMovieAndConnect(self, idname):
movie = self._imdb.get_title_by_id(idname)
self._add_node(movie.title, 'movie')
self._add_node(movie.year, 'year')
logging.info("Loading {0}".format(idname))
for genre in movie.genres:
self._add_node(genre, 'genre')
self._graph.add_edge(movie.title, genre)
for person in movie.credits:
self._add_node(person.name, 'actor')
self._graph.add_edge(movie.title, person.name, weight=movie.rating + movie.votes, rating=movie.rating, votes=movie.votes)
for person in movie.cast_summary:
self._add_node(person.name, "actor")
self._graph.add_edge(movie.title, person.name)
def addPopular(self, limit=2):
''' Add popular movies and shows
'''
shows = self._imdb.popular_shows()
#movies = self._imdb.top_250()
if limit > len(shows):
limit = len(shows)
for show in shows[:limit]:
self.addMovie(show['tconst'])
def removeNode(self, nodename):
self._graph.remove_node(nodename)
def addEdge(self, innode, outnode, prop=None):
if innode not in self._graph:
raise Exception("{0} not in graph".format(innode))
if outnode not in self._graph:
raise Exception("{0} not in graph".format(outnode))
self._graph.add_edge(innode,outnode, prop=prop)
def components(self):
comp = nx.connected_components(self._graph)
degree = nx.degree(self._graph)
def avg_degree(self):
''' Return average number of degree for each node
'''
return nx.average_neighbor_degree(self._graph)
def avg_degree_connectivity(self):
return nx.average_degree_connectivity(self._graph)
def clustering(self):
''' Compute a bipartite clustering coefficient for nodes.
'''
return nx.clustering(self._graph)
def get_item(self, item):
''' Getting node from the graph
'''
return self._graph.node[item]
def filter_edges(self, param, func):
for n, nbrs in self._graph.adjacency_iter():
for nbr, attr in nbrs.items():
if len(attr) == 0 or param not in attr: continue
data = attr[param]
if func(data): yield (n, nbr, data)
def cliques(self):
''' return all cluques from the graph
'''
return nx.find_cliques(self._graph)
def stat(self):
''' Return basic statistics of the graph
'''
return {'nodes': self._graph.number_of_nodes(), 'edges': self._graph.number_of_edges(), 'density': nx.density(self._graph)}
def save(self, outpath):
''' save graph to the file
'''
pass
class IMDBGraph:
def __init__(self, anonymize=True):
self._imdb = Imdb(anonymize=anonymize, cache=True)
self._graph = nx.Graph()
def _add_node(self, name, nodetype):
''' Add simple node without attributes
'''
if name not in self._graph.nodes():
self._graph.add_node(name, node=nodetype)
def addPerson(self, idname):
''' add New actor/actress no the graph
'''
actor = self._imdb.get_person_by_id(idname)
self._graph.add_node(actor.name)
def addMovie(self, idname):
movie = self._imdb.get_title_by_id(idname)
self._add_node(movie.title, 'movie')
logging.info("Loading {0}".format(idname))
def addMovieAndConnect(self, idname):
movie = self._imdb.get_title_by_id(idname)
self._add_node(movie.title, 'movie')
self._add_node(movie.year, 'year')
logging.info("Loading {0}".format(idname))
for genre in movie.genres:
self._add_node(genre, 'genre')
self._graph.add_edge(movie.title, genre)
for person in movie.credits:
self._add_node(person.name, 'actor')
self._graph.add_edge(movie.title,
person.name,
weight=movie.rating + movie.votes,
rating=movie.rating,
votes=movie.votes)
for person in movie.cast_summary:
self._add_node(person.name, "actor")
self._graph.add_edge(movie.title, person.name)
def addPopular(self, limit=2):
''' Add popular movies and shows
'''
shows = self._imdb.popular_shows()
#movies = self._imdb.top_250()
if limit > len(shows):
limit = len(shows)
for show in shows[:limit]:
self.addMovie(show['tconst'])
def removeNode(self, nodename):
self._graph.remove_node(nodename)
def addEdge(self, innode, outnode, prop=None):
if innode not in self._graph:
raise Exception("{0} not in graph".format(innode))
if outnode not in self._graph:
raise Exception("{0} not in graph".format(outnode))
self._graph.add_edge(innode, outnode, prop=prop)
def components(self):
comp = nx.connected_components(self._graph)
degree = nx.degree(self._graph)
def avg_degree(self):
''' Return average number of degree for each node
'''
return nx.average_neighbor_degree(self._graph)
def avg_degree_connectivity(self):
return nx.average_degree_connectivity(self._graph)
def clustering(self):
''' Compute a bipartite clustering coefficient for nodes.
'''
return nx.clustering(self._graph)
def get_item(self, item):
''' Getting node from the graph
'''
return self._graph.node[item]
def filter_edges(self, param, func):
for n, nbrs in self._graph.adjacency_iter():
for nbr, attr in nbrs.items():
if len(attr) == 0 or param not in attr: continue
data = attr[param]
if func(data): yield (n, nbr, data)
def cliques(self):
''' return all cluques from the graph
'''
return nx.find_cliques(self._graph)
def stat(self):
''' Return basic statistics of the graph
'''
return {
'nodes': self._graph.number_of_nodes(),
'edges': self._graph.number_of_edges(),
'density': nx.density(self._graph)
}
def save(self, outpath):
''' save graph to the file
'''
pass
from imdbpie import Imdb
import pandas as pd
from halo import Halo
imdb = Imdb(anonymize=True)
series = imdb.popular_shows()
cols = [
"Title", "Actors", "Genres", "Rating", "Running Time", "Year",
"Certification", "Writers"
]
df = pd.DataFrame(columns=cols)
spinner = Halo(text='Loading', spinner='dots')
spinner.start()
for j, el in enumerate(series):
show = imdb.get_title_by_id(el["tconst"])
title = show.title
actors = ', '.join(i["name"] for i in el["principals"])
genres = ', '.join(i for i in show.genres)
genres = ', '.join(i for i in show.genres)
rating = show.rating
rtime = show.runtime
year = show.year
cert = show.certification
writers = ', '.join(i.name for i in show.writers_summary)
spinner.text = "Running - " + str((j + 1) / 0.5) + "%"
