Python imdb.ImdbParser类代码示例

 def test_no_year(self):
     # Make sure parser doesn't crash for movies with no year
     parser = ImdbParser()
     parser.parse('tt3303790')
     assert parser.name == 'Master of None'
     # There is no year
     assert not parser.year

 def test_no_plot(self):
     # Make sure parser doesn't crash for movies with no plot
     parser = ImdbParser()
     parser.parse('tt0245062')
     assert parser.name == 'The Magnet'
     # There is no plot
     assert not parser.plot_outline

 def test_plot_with_links(self):
     """Make sure plot doesn't terminate at the first link. GitHub #756"""
     parser = ImdbParser()
     parser.parse('tt2503944')
     assert parser.plot_outline == ("Chef Adam Jones (Bradley Cooper) had it all - and lost it. A two-star Michelin "
                                    "rockstar with the bad habits to match, the former enfant terrible of the Paris "
                                    "restaurant scene did everything different every time out, and only ever cared "
                                    "about the thrill of creating explosions of taste. To land his own kitchen and "
                                    "that third elusive Michelin star though, he'll need the best of the best on "
                                    "his side, including the beautiful Helene (Sienna Miller).")

    def _parse_new_movie(self, imdb_url, session):
        """
        Get Movie object by parsing imdb page and save movie into the database.

        :param imdb_url: IMDB url
        :param session: Session to be used
        :return: Newly added Movie
        """
        parser = ImdbParser()
        parser.parse(imdb_url)
        # store to database
        movie = Movie()
        movie.photo = parser.photo
        movie.title = parser.name
        movie.original_title = parser.original_name
        movie.score = parser.score
        movie.votes = parser.votes
        movie.meta_score = parser.meta_score
        movie.year = parser.year
        movie.mpaa_rating = parser.mpaa_rating
        movie.plot_outline = parser.plot_outline
        movie.url = imdb_url
        for name in parser.genres:
            genre = session.query(Genre).filter(Genre.name == name).first()
            if not genre:
                genre = Genre(name)
            movie.genres.append(genre)  # pylint:disable=E1101
        for index, name in enumerate(parser.languages):
            language = session.query(Language).filter(Language.name == name).first()
            if not language:
                language = Language(name)
            movie.languages.append(MovieLanguage(language, prominence=index))
        for imdb_id, name in parser.actors.items():
            actor = session.query(Actor).filter(Actor.imdb_id == imdb_id).first()
            if not actor:
                actor = Actor(imdb_id, name)
            movie.actors.append(actor)  # pylint:disable=E1101
        for imdb_id, name in parser.directors.items():
            director = session.query(Director).filter(Director.imdb_id == imdb_id).first()
            if not director:
                director = Director(imdb_id, name)
            movie.directors.append(director)  # pylint:disable=E1101
        for imdb_id, name in parser.writers.items():
            writer = session.query(Writer).filter(Writer.imdb_id == imdb_id).first()
            if not writer:
                writer = Writer(imdb_id, name)
            movie.writers.append(writer)  # pylint:disable=E1101
            # so that we can track how long since we've updated the info later
        movie.updated = datetime.now()
        session.add(movie)
        return movie

 def test_parsed_data(self):
     parser = ImdbParser()
     parser.parse('tt0114814')
     assert parser.actors == {
         'nm0000592': 'Pete Postlethwaite',
         'nm0261452': 'Christine Estabrook',
         'nm0000751': 'Suzy Amis',
         'nm0000286': 'Stephen Baldwin',
         'nm0000445': 'Dan Hedaya',
         'nm0800339': 'Phillipe Simon',
         'nm0002064': 'Giancarlo Esposito',
         'nm0001590': 'Chazz Palminteri',
         'nm0000321': 'Gabriel Byrne',
         'nm0790436': 'Jack Shearer',
         'nm0000228': 'Kevin Spacey',
         'nm0001629': 'Kevin Pollak',
         'nm0107808': 'Carl Bressler',
         'nm0001125': 'Benicio Del Toro',
         'nm0000860': 'Paul Bartel'
     }, 'Actors not parsed correctly'
     assert parser.directors == {'nm0001741': 'Bryan Singer'}, 'Directors not parsed correctly'
     assert len(set(parser.genres).intersection([u'crime', u'drama', u'mystery', u'thriller'])) == \
         len([u'crime', u'drama', u'mystery', u'thriller']), 'Genres not parsed correctly'
     assert parser.imdb_id == 'tt0114814', 'ID not parsed correctly'
     assert len(set(parser.languages).intersection(
         ['english', 'hungarian', 'spanish', 'french'])) == 4, 'Languages not parsed correctly'
     assert parser.mpaa_rating == 'R', 'Rating not parsed correctly'
     assert parser.name == 'The Usual Suspects', 'Name not parsed correctly'
     assert (parser.photo ==
             'https://images-na.ssl-images-amazon.com/images/M/[email protected]@.'
             '_V1_UX182_CR0,0,182,268_AL_.jpg'
             ), 'Photo not parsed correctly'
     assert parser.plot_outline == (
         'Following a truck hijack in New York, five conmen are arrested and brought together for questioning. '
         'As none of them are guilty, they plan a revenge operation against the police. The operation goes well, '
         'but then the influence of a legendary mastermind criminal called Keyser S\xf6ze is felt. It becomes '
         'clear that each one of them has wronged S\xf6ze at some point and must pay back now. The payback job '
         'leaves 27 men dead in a boat explosion, but the real question arises now: Who actually is Keyser S\xf6ze?'
     ), 'Plot outline not parsed correctly'
     assert 8.0 < parser.score < 9.0, 'Score not parsed correctly'
     assert parser.url == 'http://www.imdb.com/title/tt0114814/', 'URL not parsed correctly'
     assert 400000 < parser.votes < 1000000, 'Votes not parsed correctly'
     assert parser.year == 1995, 'Year not parsed correctly'

def upgrade(ver, session):
    if ver == 0:
        # Translate old qualities into new quality requirements
        movie_table = table_schema('movie_queue', session)
        for row in session.execute(select([movie_table.c.id, movie_table.c.quality])):
            # Webdl quality no longer has dash
            new_qual = row['quality'].replace('web-dl', 'webdl')
            if new_qual.lower() != 'any':
                # Old behavior was to get specified quality or greater, approximate that with new system
                new_qual = ' '.join(qual + '+' for qual in new_qual.split(' '))
            session.execute(update(movie_table, movie_table.c.id == row['id'],
                                   {'quality': new_qual}))
        ver = 1
    if ver == 1:
        # Bad upgrade left some qualities as 'ANY+'
        movie_table = table_schema('movie_queue', session)
        for row in session.execute(select([movie_table.c.id, movie_table.c.quality])):
            if row['quality'].lower() == 'any+':
                session.execute(update(movie_table, movie_table.c.id == row['id'],
                                       {'quality': 'ANY'}))
        ver = 2
    if ver == 2:
        from flexget.utils.imdb import ImdbParser
        # Corrupted movie titles may be in the queue due to imdb layout changes. GitHub #729
        movie_table = table_schema('movie_queue', session)
        queue_base_table = table_schema('queue', session)
        query = select([movie_table.c.id, movie_table.c.imdb_id, queue_base_table.c.title])
        query = query.where(movie_table.c.id == queue_base_table.c.id)
        for row in session.execute(query):
            if row['imdb_id'] and (not row['title'] or row['title'] == 'None' or '\n' in row['title']):
                log.info('Fixing movie_queue title for %s' % row['imdb_id'])
                parser = ImdbParser()
                parser.parse(row['imdb_id'])
                if parser.name:
                    session.execute(update(queue_base_table, queue_base_table.c.id == row['id'],
                                           {'title': parser.name}))
        ver = 3
    if ver == 3:
        # adding queue_name column to movie_queue table and setting initial value to default)
        table_add_column('movie_queue', 'queue_name', Unicode, session, default='default')
        ver = 4
    return ver

def upgrade(ver, session):
    if ver == 0:
        # Translate old qualities into new quality requirements
        movie_table = table_schema("movie_queue", session)
        for row in session.execute(select([movie_table.c.id, movie_table.c.quality])):
            # Webdl quality no longer has dash
            new_qual = row["quality"].replace("web-dl", "webdl")
            if new_qual.lower() != "any":
                # Old behavior was to get specified quality or greater, approximate that with new system
                new_qual = " ".join(qual + "+" for qual in new_qual.split(" "))
            session.execute(update(movie_table, movie_table.c.id == row["id"], {"quality": new_qual}))
        ver = 1
    if ver == 1:
        # Bad upgrade left some qualities as 'ANY+'
        movie_table = table_schema("movie_queue", session)
        for row in session.execute(select([movie_table.c.id, movie_table.c.quality])):
            if row["quality"].lower() == "any+":
                session.execute(update(movie_table, movie_table.c.id == row["id"], {"quality": "ANY"}))
        ver = 2
    if ver == 2:
        from flexget.utils.imdb import ImdbParser

        # Corrupted movie titles may be in the queue due to imdb layout changes. GitHub #729
        movie_table = table_schema("movie_queue", session)
        queue_base_table = table_schema("queue", session)
        query = select([movie_table.c.id, movie_table.c.imdb_id, queue_base_table.c.title])
        query = query.where(movie_table.c.id == queue_base_table.c.id)
        for row in session.execute(query):
            if row["imdb_id"] and (not row["title"] or row["title"] == "None" or "\n" in row["title"]):
                log.info("Fixing movie_queue title for %s" % row["imdb_id"])
                parser = ImdbParser()
                parser.parse(row["imdb_id"])
                if parser.name:
                    session.execute(
                        update(queue_base_table, queue_base_table.c.id == row["id"], {"title": parser.name})
                    )
        ver = 3
    return ver

#.........这里部分代码省略.........
                    session.add(result)
                    log.verbose('Found %s' % (entry['imdb_url']))
                else:
                    log_once('Imdb lookup failed for %s' % entry['title'], log)
                    # store FAIL for this title
                    result = SearchResult(entry['title'])
                    result.fails = True
                    session.add(result)
                    raise PluginError('Title `%s` lookup failed' % entry['title'])


            # check if this imdb page has been parsed & cached
            movie = session.query(Movie).\
                options(joinedload_all(Movie.genres, Movie.languages,
                Movie.actors, Movie.directors)).\
                filter(Movie.url == entry['imdb_url']).first()

            refresh_interval = 2
            if movie:
                if movie.year:
                    age = (datetime.now().year - movie.year)
                    refresh_interval += age * 5
                    log.debug('cached movie `%s` age %i refresh interval %i days' % (movie.title, age, refresh_interval))

            if not movie or movie.updated is None or \
               movie.updated < datetime.now() - timedelta(days=refresh_interval):
                # Remove the old movie, we'll store another one later.
                session.query(Movie).filter(Movie.url == entry['imdb_url']).delete()
                # search and store to cache
                if 'title' in entry:
                    log.verbose('Parsing imdb for `%s`' % entry['title'])
                else:
                    log.verbose('Parsing imdb for `%s`' % entry['imdb_id'])
                try:
                    take_a_break = True
                    imdb = ImdbParser()
                    imdb.parse(entry['imdb_url'])
                    # store to database
                    movie = Movie()
                    movie.photo = imdb.photo
                    movie.title = imdb.name
                    movie.score = imdb.score
                    movie.votes = imdb.votes
                    movie.year = imdb.year
                    movie.mpaa_rating = imdb.mpaa_rating
                    movie.plot_outline = imdb.plot_outline
                    movie.url = entry['imdb_url']
                    for name in imdb.genres:
                        genre = session.query(Genre).\
                            filter(Genre.name == name).first()
                        if not genre:
                            genre = Genre(name)
                        movie.genres.append(genre) # pylint:disable=E1101
                    for name in imdb.languages:
                        language = session.query(Language).\
                            filter(Language.name == name).first()
                        if not language:
                            language = Language(name)
                        movie.languages.append(language) # pylint:disable=E1101
                    for imdb_id, name in imdb.actors.iteritems():
                        actor = session.query(Actor).\
                            filter(Actor.imdb_id == imdb_id).first()
                        if not actor:
                            actor = Actor(imdb_id, name)
                        movie.actors.append(actor) # pylint:disable=E1101
                    for imdb_id, name in imdb.directors.iteritems():
                        director = session.query(Director).\
                            filter(Director.imdb_id == imdb_id).first()
                        if not director:
                            director = Director(imdb_id, name)
                        movie.directors.append(director) # pylint:disable=E1101
                    # so that we can track how long since we've updated the info later
                    movie.updated = datetime.now()
                    session.add(movie)

                except UnicodeDecodeError:
                    log.error('Unable to determine encoding for %s. Installing chardet library may help.' % entry['imdb_url'])
                    # store cache so this will not be tried again
                    movie = Movie()
                    movie.url = entry['imdb_url']
                    session.add(movie)
                    raise PluginError('UnicodeDecodeError')
                except ValueError, e:
                    # TODO: might be a little too broad catch, what was this for anyway? ;P
                    if manager.options.debug:
                        log.exception(e)
                    raise PluginError('Invalid parameter: %s' % entry['imdb_url'], log)

            for att in ['title', 'score', 'votes', 'year', 'genres', 'languages', 'actors', 'directors', 'mpaa_rating']:
                log.trace('movie.%s: %s' % (att, getattr(movie, att)))

            # store to entry
            entry.update_using_map(self.field_map, movie)

            # give imdb a little break between requests (see: http://flexget.com/ticket/129#comment:1)
            if (take_a_break and
                not manager.options.debug and
                not manager.unit_test):
                import time
                time.sleep(3)

#.........这里部分代码省略.........
                        try:
                            match = tvdbapi[series_name]
                            logger.debug("Show found")
                            instance.tvdbid_id = int(match['id'])

                            if match['imdb_id'] is not None:
                                logger.debug("also found imdbid %s from thetvdb" % match['imdb_id'])
                                instance.imdbid_id = int(match['imdb_id'].lstrip("tt"))
                        except tvdb_shownotfound:
                            logger.exception("Error finding show on thetvdb %s" % series_name)
                        except Exception as e:
                            logger.exception("Error finding : %s via thetvdb.com due to  %s" % (series_name, e.message))
                else:
                    logger.exception("Unable to parse series info")

        else:
            #must be a movie!
            if instance.imdbid_id is None:
                logger.debug("Looks like we are working with a Movie")
                #Lets try find the movie details
                parser = instance.metaparser()

                movie_title = parser.details['title']

                if 'year' in parser.details:
                    movie_year = parser.details['year']
                else:
                    movie_year = None

                imdbs = ImdbSearch()
                results = imdbs.best_match(movie_title, movie_year)

                if results and results['match'] > 0.70:
                    movieObj = ImdbParser()
                    movieObj.parse(results['url'])

                    logger.debug("Found imdb movie id %s" % movieObj.imdb_id)

                    instance.imdbid_id = int(movieObj.imdb_id.lstrip("tt"))
                else:
                    logger.debug("Didnt find a good enough match on imdb")

        #Now we have sorted both imdbid and thetvdbid lets sort it all out

        #If we have a tvdbid do we need to add it to the db or does it exist or ignored?
        if instance.tvdbid_id is not None and instance.tvdbid_id != "":

            #Does it already exist?
            try:
                if instance.tvdbid:
                    #Do we need to update it
                    curTime = timezone.now()
                    hours = 0

                    if instance.tvdbid.updated is None:
                        hours = 50
                    else:
                        diff = curTime - instance.tvdbid.updated
                        hours = diff.total_seconds() / 60 / 60

                    if hours > 24:
                        try:
                            instance.tvdbid.update_from_tvdb()
                            instance.tvdbid.save()
                        except Exception as e:
                            logger.exception("Error updating TVDB info %s" % e.message)