IMDb: Page not found

([\d\.]+)', 'type': 'float', }, 'budget': { 'page': 'business', 're': [ '

Budget

\s*?\$(.*?).*?>(.*?).*?(.*?)', lambda ll: [strip_tags(l) for l in ll] ], 'type': 'list' }, 'cinematographer': { 'page': 'combined', 're': [ lambda data: data.split('Series Crew')[0], 'Cinematography by(.*?)', '(.*?)' ], 'type': 'list' }, 'connections': { 'page': 'trivia?tab=mc', 're': '

(.*?)

(.*?)(<\/div>\n

Country:

.*?

', #'(.*?)', #links changed to work with existing caches, just take all links '(.*?)', ], 'type': 'list' }, 'creator': { 'page': 'combined', 're': [ '

Creator.?:

.*?

(.*?)

', '(.*?)' ], 'type': 'list' }, '_director': { 'page': 'combined', 're': [ '

Director:

.*?

(.*?)

', '(.*?)' ], 'type': 'list' }, 'composer': { 'page': 'combined', 're': [ lambda data: data.split('Series Crew')[0], 'Original Music by(.*?)', '.*?(.*?)', 'type': 'string' }, 'filmingLocations': { 'page': 'locations', 're': [ '(.*?)', lambda data: data.strip(), ], 'type': 'list' }, 'genre': { 'page': 'combined', 're': [ '

Genre:

(.*?)(.*?)' ], 'type': 'list' }, 'gross': { 'page': 'business', 're': [ '

Gross

\s*?\$(.*?)

Language:

.*?

', #'(.*?)', #links changed to work with existing caches, just take all links '(.*?)', ], 'type': 'list' }, 'summary': { 'page': 'plotsummary', 're': '

.*?

(.*?)', 'type': 'string' }, 'posterId': { 'page': 'combined', 're': '/primary-photo/media/rm(.*?)/tt', 'type': 'string' }, 'posterIds': { 'page': 'posters', 're': '/unknown-thumbnail/media/rm(.*?)/tt', 'type': 'list' }, 'producer': { 'page': 'combined', 're': [ lambda data: data.split('Series Crew')[0], 'Produced by(.*?)', '(.*?)' ], 'type': 'list' }, 'productionCompany': { 'page': 'combined', 're': [ 'Production Companies

(.*?)

', '(.*?)' ], 'type': 'list' }, 'rating': { 'page': 'combined', 're': '
.*?([\d,.]+?)/10', 'type': 'float' }, 'releasedate': { 'page': 'releaseinfo', 're': [ '(.*?)', ox.strip_tags, ], 'type': 'list' }, 'reviews': { 'page': 'externalreviews', 're': [ '
(.*?)
', '
(.*?)
' ], 'type': 'list' }, 'runtime': { 'page': 'combined', 're': '
Runtime:
.*?([0-9]+ sec|[0-9]+ min).*?
', 'type': 'string' }, 'color': { 'page': 'combined', 're': [ '
Color:
(.*?)
', '(.*?)' ], 'type': 'list' }, 'sound': { 'page': 'combined', 're': [ '
Sound Mix:
(.*?)
', '(.*?)' ], 'type': 'list' }, 'season': { 'page': 'combined', 're': [ '
Original Air Date:
.*?
(.*?)
', '$Season (\d+), Episode \d+$', ], 'type': 'int' }, 'episode': { 'page': 'combined', 're': [ '
Original Air Date:
.*?
(.*?)
', '$Season \d+, Episode (\d+)$', ], 'type': 'int' }, 'series': { 'page': 'combined', 're': '
TV Series:
.*?(TV series|TV mini-series) ', 'type': 'string' }, 'title': { 'page': 'combined', 're': '
(.*?) ', 'type': 'string' }, 'trivia': { 'page': 'trivia', 're': '
(.*?)([\d,]*?) votes', 'type': 'string' }, 'writer': { 'page': 'combined', 're': [ lambda data: data.split('Series Crew')[0], 'Writing credits(.*?)', '(.*?)' ], 'type': 'list' }, 'year': { 'page': 'combined', 're': '="og:title" content=".*?\((\d{4}).*?"', 'type': 'int' } } def read_url(self, url, timeout): if not url in self._cache: self._cache[url] = read_url(url, timeout=timeout, unicode=True) return self._cache[url] def __init__(self, id, timeout=-1): #use akas.imdb.com to always get original title: #http://www.imdb.com/help/show_leaf?titlelanguagedisplay self.baseUrl = "http://akas.imdb.com/title/tt%s/" % id super(Imdb, self).__init__(timeout) url = self.baseUrl + 'combined' page = self.read_url(url, timeout=-1) if 'IMDb: Page not found' in page \ or 'The requested URL was not found on our server.' in page: return if "
We're sorry, something went wrong.
" in page: time.sleep(1) super(Imdb, self).__init__(0) if 'alternativeTitles' in self: if len(self['alternativeTitles']) == 2 and \ isinstance(self['alternativeTitles'][0], basestring): self['alternativeTitles'] = [self['alternativeTitles']] types = {} stop_words = [ 'alternative spelling', 'alternative title', 'alternative transliteration', 'complete title', 'IMAX version', 'informal short title', 'Japan (imdb display title)', 'longer version', 'new title', 'recut version', 'reissue title', 'restored version', 'script title', 'TV title', 'working title', 'pre-release title', 'promotional abbreviation', 'short title', '(subtitle)', 'closing credits title', ] #ignore english japanese titles #for movies that are not only from japan if ['Japan'] != self.get('country', []): stop_words += [ 'Japan (English title)' ] for t in self.get('alternativeTitles', []): for type in t[0].split('/'): type = type.strip() stop_word = False for key in stop_words: if key in type: stop_word = True break if not stop_word and not type in types: types[type] = t[1] regexps = [ "^.+ $imdb display title$ $English title$$", "^USA $imdb display title$$", "^International $English title$$", "^UK $imdb display title$$", "^International $.+$ $English title$$", ] if 'Hong Kong' in self.get('country', []): regexps += [ "Hong Kong $English title$" ] english_countries = ('USA', 'UK', 'Australia', 'New Zealand') if not filter(lambda c: c in english_countries, self.get('country', [])): regexps += [ "^[^(]+ $English title$$", "^.+ $.+$ $English title$$", "^USA$", "^UK$", "^USA $.+$$", "^UK $.+$$", "^Australia $.+$$", "(literal English title)", "^International $.+ title$$", "^International $.+$ $.+ title$$", ] for regexp in regexps: for type in types: if re.compile(regexp).findall(type): #print types[type], type self['internationalTitle'] = types[type] break if 'internationalTitle' in self: break def cleanup_title(title): if title.startswith('"') and title.endswith('"'): title = title[1:-1] if title.startswith("'") and title.endswith("'"): title = title[1:-1] title = re.sub('$\#[.\d]+$', '', title) return title.strip() for t in ('title', 'internationalTitle'): if t in self: self[t] = cleanup_title(self[t]) if 'internationalTitle' in self and \ self.get('title', '').lower() == self['internationalTitle'].lower(): del self['internationalTitle'] if 'alternativeTitles' in self: alt = {} for t in self['alternativeTitles']: title = cleanup_title(t[1]) if title not in (self.get('title'), self.get('internationalTitle')): if title not in alt: alt[title] = [] for c in t[0].split('/'): c = c.replace('International', '').split('(')[0].strip() if c: alt[title].append(c) self['alternativeTitles'] = [] for t in sorted(alt, lambda a, b: cmp(sorted(alt[a]), sorted(alt[b]))): if alt[t]: self['alternativeTitles'].append((t, sorted(alt[t]))) if not self['alternativeTitles']: del self['alternativeTitles'] if 'internationalTitle' in self: self['originalTitle'] = self['title'] self['title'] = self.pop('internationalTitle') if 'runtime' in self and self['runtime']: if 'min' in self['runtime']: base=60 else: base=1 self['runtime'] = int(find_re(self['runtime'], '([0-9]+)')) * base if 'runtime' in self and not self['runtime']: del self['runtime'] if 'votes' in self: self['votes'] = self['votes'].replace(',', '') if 'cast' in self: if isinstance(self['cast'][0], basestring): self['cast'] = [self['cast']] self['actor'] = [c[0] for c in self['cast']] def cleanup_character(c): c = c.replace('(uncredited)', '').strip() return c self['cast'] = [{'actor': x[0], 'character': cleanup_character(x[1])} for x in self['cast']] if 'connections' in self: cc={} if len(self['connections']) == 3 and isinstance(self['connections'][0], basestring): self['connections'] = [self['connections']] for rel, data, _ in self['connections']: #cc[unicode(rel)] = re.compile('(.*?)').findall(data) def get_conn(c): r = { 'id': c[0], 'title': cleanup_title(c[1]), } description = c[2].split('
') if len(description) == 2 and description[-1].strip() != '-': r['description'] = description[-1].strip() return r cc[unicode(rel)] = map(get_conn, re.compile('(.*?)(.*?)<\/div', re.DOTALL).findall(data)) self['connections'] = cc for key in ('country', 'genre'): if key in self: self[key] = filter(lambda x: x.lower() != 'home', self[key]) #0092999 if '_director' in self: if 'series' in self or 'isSeries' in self: self['creator'] = self.pop('_director') else: del self['_director'] if 'isSeries' in self: del self['isSeries'] self['isSeries'] = True if 'episodeTitle' in self: self['episodeTitle'] = re.sub('Episode \#\d+\.\d+', '', self['episodeTitle']) if 'series' in self: series = Imdb(self['series'], timeout=timeout) self['seriesTitle'] = series['title'] if 'episodeTitle' in self: self['seriesTitle'] = series['title'] if 'season' in self and 'episode' in self: self['title'] = "%s (S%02dE%02d) %s" % ( self['seriesTitle'], self['season'], self['episode'], self['episodeTitle']) else: self['title'] = "%s (S01) %s" % (self['seriesTitle'], self['episodeTitle']) self['season'] = 1 self['title'] = self['title'].strip() if 'director' in self: self['episodeDirector'] = self['director'] if not 'creator' in series and 'director' in series: series['creator'] = series['director'] if len(series['creator']) > 10: series['creator'] = series['director'][:1] for key in ['creator', 'country']: if key in series: self[key] = series[key] if 'year' in series: self['seriesYear'] = series['year'] if not 'year' in self: self['year'] = series['year'] if 'year' in self: self['episodeYear'] = self['year'] if 'creator' in self: self['seriesDirector'] = self['creator'] if 'originalTitle' in self: del self['originalTitle'] else: for key in ('seriesTitle', 'episodeTitle', 'season', 'episode'): if key in self: del self[key] if 'creator' in self: if 'director' in self: self['episodeDirector'] = self['director'] self['director'] = self['creator'] #make lists unique but keep order for key in ('director', 'language'): if key in self: self[key] = [x for i,x in enumerate(self[key]) if x not in self[key][i+1:]] for key in ('actor', 'writer', 'producer', 'editor', 'composer'): if key in self: if isinstance(self[key][0], list): self[key] = [i[0] for i in self[key] if i] self[key] = sorted(list(set(self[key])), lambda a, b: self[key].index(a) - self[key].index(b)) if 'budget' in self and 'gross' in self: self['profit'] = self['gross'] - self['budget'] if 'releasedate' in self: def parse_date(d): try: d = datetime.strptime(d, '%d %B %Y') except: try: d = datetime.strptime(d, '%B %Y') except: return 'x' return '%d-%02d-%02d' % (d.year, d.month, d.day) self['releasedate'] = min([ parse_date(d) for d in self['releasedate'] ]) if self['releasedate'] == 'x': del self['releasedate'] if 'summary' in self: self['summary'] = self['summary'].split('>> get_movie_by_title(u'"Father Knows Best" (1954) {(#5.34)}') u'1602860' >>> get_movie_by_title(u'The Matrix (1999)') u'0133093' >>> get_movie_by_title(u'Little Egypt (1951)') u'0043748' >>> get_movie_by_title(u'Little Egypt (1897/I)') u'0214882' >>> get_movie_by_title(u'Little Egypt') None >>> get_movie_by_title(u'"Dexter" (2006) {Father Knows Best (#1.9)}') u'0866567' ''' params = {'s':'tt','q': title} if isinstance(title, unicode): try: params['q'] = unicodedata.normalize('NFKC', params['q']).encode('latin-1') except: params['q'] = params['q'].encode('utf-8') params = urllib.urlencode(params) url = "http://akas.imdb.com/find?" + params data = read_url(url, timeout=timeout, unicode=True) #if search results in redirect, get id of current page r = '' results = re.compile(r).findall(data) if results: return results[0] return None def get_movie_id(title, director='', year='', timeout=-1): ''' >>> get_movie_id('The Matrix') u'0133093' >>> get_movie_id('2 or 3 Things I Know About Her', 'Jean-Luc Godard') u'0060304' >>> get_movie_id('2 or 3 Things I Know About Her', 'Jean-Luc Godard', '1967') u'0060304' >>> get_movie_id(u"Histoire(s) du cinema: Le controle de l'univers", 'Jean-Luc Godard') u'0179214' >>> get_movie_id(u"Histoire(s) du cinéma: Le contrôle de l'univers", 'Jean-Luc Godard') u'0179214' ''' imdbId = { (u'Le jour se l\xe8ve', u'Marcel Carn\xe9'): '0031514', (u'Wings', u'Larisa Shepitko'): '0061196', (u'The Ascent', u'Larisa Shepitko'): '0075404', (u'Fanny and Alexander', u'Ingmar Bergman'): '0083922', (u'Torment', u'Alf Sj\xf6berg'): '0036914', (u'Crisis', u'Ingmar Bergman'): '0038675', (u'To Joy', u'Ingmar Bergman'): '0043048', (u'Humain, trop humain', u'Louis Malle'): '0071635', (u'Place de la R\xe9publique', u'Louis Malle'): '0071999', (u'God\u2019s Country', u'Louis Malle'): '0091125', (u'Flunky, Work Hard', u'Mikio Naruse'): '0022036', (u'The Courtesans of Bombay', u'Richard Robbins') : '0163591', (u'Je tu il elle', u'Chantal Akerman') : '0071690', (u'Hotel Monterey', u'Chantal Akerman') : '0068725', (u'No Blood Relation', u'Mikio Naruse') : '023261', (u'Apart from You', u'Mikio Naruse') : '0024214', (u'Every-Night Dreams', u'Mikio Naruse') : '0024793', (u'Street Without End', u'Mikio Naruse') : '0025338', (u'Sisters of the Gion', u'Kenji Mizoguchi') : '0027672', (u'Osaka Elegy', u'Kenji Mizoguchi') : '0028021', (u'Blaise Pascal', u'Roberto Rossellini') : '0066839', (u'Japanese Girls at the Harbor', u'Hiroshi Shimizu') : '0160535', (u'The Private Life of Don Juan', u'Alexander Korda') : '0025681', (u'Last Holiday', u'Henry Cass') : '0042665', (u'A Colt Is My Passport', u'Takashi Nomura') : '0330536', (u'Androcles and the Lion', u'Chester Erskine') : '0044355', (u'Major Barbara', u'Gabriel Pascal') : '0033868', (u'Come On Children', u'Allan King') : '0269104', (u'Jimi Plays Monterey & Shake! Otis at Monterey', u'D. A. Pennebaker and Chris Hegedus') : '', (u'Martha Graham: Dance on Film', u'Nathan Kroll') : '', (u'Carmen', u'Carlos Saura'): '0085297', (u'The Story of a Cheat', u'Sacha Guitry'): '0028201', (u'Weekend', 'Andrew Haigh'): '1714210', }.get((title, director), None) if imdbId: return imdbId params = {'s':'tt','q': title} if director: params['q'] = u'"%s" %s' % (title, director) if year: params['q'] = u'"%s (%s)" %s' % (title, year, director) google_query = "site:imdb.com %s" % params['q'] if isinstance(params['q'], unicode): try: params['q'] = unicodedata.normalize('NFKC', params['q']).encode('latin-1') except: params['q'] = params['q'].encode('utf-8') params = urllib.urlencode(params) url = "http://akas.imdb.com/find?" + params #print url data = read_url(url, timeout=timeout, unicode=True) #if search results in redirect, get id of current page r = '' results = re.compile(r).findall(data) if results: return results[0] #otherwise get first result r = '.*?>> get_movie_poster('0133093') 'http://ia.media-imdb.com/images/M/MV5BMjEzNjg1NTg2NV5BMl5BanBnXkFtZTYwNjY3MzQ5._V1._SX338_SY475_.jpg' >>> get_movie_poster('0994352') 'http://ia.media-imdb.com/images/M/MV5BMjA3NzMyMzU1MV5BMl5BanBnXkFtZTcwNjc1ODUwMg@@._V1._SX594_SY755_.jpg' ''' info = ImdbCombined(imdbId) if 'posterId' in info: url = "http://www.imdb.com/rg/action-box-title/primary-photo/media/rm%s/tt%s" % (info['posterId'], imdbId) data = read_url(url) poster = find_re(data, 'img id="primary-img".*?src="(.*?)"') return poster elif 'series' in info: return get_movie_poster(info['series']) return '' def get_episodes(imdbId, season=None): episodes = {} url = 'http://www.imdb.com/title/tt%s/episodes' % imdbId if season: url += '?season=%d' % season data = ox.cache.read_url(url) for e in re.compile('
.*?
S(\d+), Ep(\d+)<\/div>\n<\/div>', re.DOTALL).findall(data): episodes['S%02dE%02d' %(int(e[1]), int(e[2]))] = e[0] else: data = ox.cache.read_url(url) match = re.compile('Season (\d+)').findall(data) if match: for season in range(1, int(match[0]) + 1): episodes.update(get_episodes(imdbId, season)) return episodes def max_votes(): url = 'http://www.imdb.com/search/title?num_votes=500000,&sort=num_votes,desc' data = ox.cache.read_url(url) votes = max([int(v.replace(',', '')) for v in re.compile('([\d,]+)').findall(data)]) return votes def guess(title, director='', timeout=-1): return get_movie_id(title, director, timeout=timeout) if __name__ == "__main__": import json print json.dumps(Imdb('0306414'), indent=2) #print json.dumps(Imdb('0133093'), indent=2)