# -*- coding: utf-8 -*- # vi:si:et:sw=4:sts=4:ts=4 import urllib2 from urllib import quote, unquote import re import os import time import ox from ox import findRe, stripTags from ox.normalize import normalizeTitle, normalizeImdbId from siteparser import SiteParser import google class Imdb(SiteParser): regex = { 'alternative_titles': { 'page': 'releaseinfo', 're': [ 'name="akas".*?(.*?)', "td>(.*?)\n\n(.*?)" ], 'type': 'list' }, 'cast': { 'page': 'combined', 're': [ '.*?>(.*?).*?(.*?)', lambda ll: [stripTags(l) for l in ll] ], 'type': 'list' }, 'cinematographers': { 'page': 'combined', 're': [ lambda data: data.split('Series Crew')[0], 'Cinematography by(.*?)', '(.*?)' ], 'type': 'list' }, 'connections': { 'page': 'movieconnections', 're': '
(.*?)
(.*?)\n\n', 'type': 'list' }, 'countries': { 'page': 'combined', 're': '(.*?)', 'type': 'list' }, 'directors': { 'page': 'combined', 're': [ lambda data: data.split('Series Crew')[0], 'Directed by(.*?)', '(.*?)' ], 'type': 'list' }, 'editors': { 'page': 'combined', 're': [ lambda data: data.split('Series Crew')[0], 'Film Editing by(.*?)', '(.*?)' ], 'type': 'list' }, 'episode_title': { 'page': 'combined', 're': '
.*?(.*?)', 'type': 'string' }, 'filming_locations': { 'page': 'locations', 're': '(.*?)', 'type': 'list' }, 'genres': { 'page': 'combined', 're': '(.*?)', 'type': 'list' }, 'keywords': { 'page': 'keywords', 're': '(.*?)', 'type': 'list' }, 'languages': { 'page': 'combined', 're': '(.*?)', 'type': 'list' }, 'original_title': { 'page': 'combined', 're': '(.*?) (original title)', 'type': 'string' }, 'plot': { 'page': 'plotsummary', 're': '
.*?

(.*?)', 'type': 'string' }, 'poster_id': { 'page': 'combined', 're': '/primary-photo/media/rm(.*?)/tt', 'type': 'list' }, 'poster_ids': { 'page': 'posters', 're': '/unknown-thumbnail/media/rm(.*?)/tt', 'type': 'list' }, 'producers': { 'page': 'combined', 're': [ lambda data: data.split('Series Crew')[0], 'Produced by(.*?)', '(.*?)' ], 'type': 'list' }, 'rating': { 'page': 'combined', 're': '

.*?([\d,.]?)/10', 'type': 'float' }, 'release date': { 'page': 'releaseinfo', 're': '.*? ', 'type': 'date' }, 'reviews': { 'page': 'externalreviews', 're': [ '
    (.*?)
', '
  • (.*?)
  • ' ], 'type': 'list' }, 'runtime': { 'page': 'combined', 're': '
    Runtime:
    .*?([0-9]+ sec|[0-9]+ min).*?
    ', 'type': 'string' }, '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:
    .*?(.*?)
    ', 'type': 'list', }, 'votes': { 'page': 'combined', 're': '
    ([\d,]*?) votes', 'type': 'string' }, 'writers': { 'page': 'combined', 're': [ lambda data: data.split('Series Crew')[0], 'Writing credits(.*?)', '(.*?)' ], 'type': 'list' }, 'year': { 'page': 'combined', 're': '').findall(data) self['connections'] = cc for key in ('countries', 'genres'): if key in self: self[key] = filter(lambda x: x.lower() != 'home', self[key]) if 'series' in self: if 'episode_title' in self: self['series_title'] = self['title'] self['title'] = "%s: %s" % (self['series_title'], self['episode_title']) if 'episode_title' in self and 'season' in self and 'episode' in self: self['title'] = "%s (S%02dE%02d) %s" % ( self['series_title'], self['season'], self['episode'], self['episode_title']) else: for key in ('series_title', 'episode_title', 'season', 'episode'): if key in self: del self[key] class ImdbCombined(Imdb): def __init__(self, id, timeout=-1): _regex = {} for key in self.regex: if self.regex[key]['page'] == 'combined': _regex[key] = self.regex[key] self.regex = _regex super(ImdbCombined, self).__init__(id, timeout) def getMovieId(title, director='', year=''): ''' >>> getMovieId('The Matrix') '0133093' ''' if year: title = "%s (%s)" % (title, year) if director: query = 'site:imdb.com %s "%s"' % (director, title) else: query = 'site:imdb.com "%s"' % title for (name, url, desc) in google.find(query, 5, timeout=-1): if url.startswith('http://www.imdb.com/title/tt'): return url[28:35] return '' def guess(title, director='', timeout=google.DEFAULT_TIMEOUT): #FIXME: proper file -> title title = title.split('-')[0] title = title.split('(')[0] title = title.split('.')[0] title = title.strip() imdb_url = 'http://www.imdb.com/find?q=%s' % quote(title.encode('utf-8')) return_url = '' #lest first try google #i.e. site:imdb.com Michael Stevens "Sin" if director: search = 'site:imdb.com %s "%s"' % (director, title) else: search = 'site:imdb.com "%s"' % title for (name, url, desc) in google.find(search, 2, timeout=timeout): if url.startswith('http://www.imdb.com/title/tt'): return normalizeImdbId(int(ox.intValue(url))) try: req = urllib2.Request(imdb_url, None, ox.net.DEFAULT_HEADERS) u = urllib2.urlopen(req) data = u.read() return_url = u.url u.close() except: return None if return_url.startswith('http://www.imdb.com/title/tt'): return return_url[28:35] if data: imdb_id = findRe(data.replace('\n', ' '), 'Popular Results.*?
    1. .*?