# coding=utf-8 import io import logging import os import time import traceback from urllib import parse import requests import inflect import re import json import html import zipfile import rarfile from babelfish import language_converters from guessit import guessit from dogpile.cache.api import NO_VALUE from subliminal import Episode, ProviderError from subliminal.exceptions import ConfigurationError, ServiceUnavailable from subliminal.utils import sanitize_release_group from subliminal.cache import region from subliminal_patch.http import RetryingCFSession from subliminal_patch.providers import Provider from subliminal_patch.providers.mixins import ProviderSubtitleArchiveMixin from subliminal_patch.subtitle import Subtitle, guess_matches from subliminal_patch.converters.subscene import language_ids, supported_languages from subscene_api.subscene import search, SearchTypes, Subtitle as APISubtitle, SITE_DOMAIN from subzero.language import Language p = inflect.engine() language_converters.register('subscene = subliminal_patch.converters.subscene:SubsceneConverter') logger = logging.getLogger(__name__) class SubsceneSubtitle(Subtitle): provider_name = 'subscene' hearing_impaired_verifiable = True is_pack = False page_link = None season = None episode = None releases = None def __init__(self, language, release_info, hearing_impaired=False, page_link=None, encoding=None, mods=None, asked_for_release_group=None, asked_for_episode=None): super(SubsceneSubtitle, self).__init__(language, hearing_impaired=hearing_impaired, page_link=page_link, encoding=encoding, mods=mods) self.release_info = self.releases = release_info self.asked_for_episode = asked_for_episode self.asked_for_release_group = asked_for_release_group self.season = None self.episode = None @classmethod def from_api(cls, s): return cls(Language.fromsubscene(s.language.strip()), s.title, hearing_impaired=s.hearing_impaired, page_link=s.url) @property def id(self): return self.page_link @property def numeric_id(self): return self.page_link.split("/")[-1] def get_matches(self, video): matches = set() if self.release_info.strip() == get_video_filename(video): logger.debug("Using hash match as the release name is the same") matches |= {"hash"} # episode if isinstance(video, Episode): guess = guessit(self.release_info, {'type': 'episode'}) self.season = guess.get("season") self.episode = guess.get("episode") matches |= guess_matches(video, guess) if "season" in matches and "episode" not in guess: # pack matches.add("episode") logger.debug("%r is a pack", self) self.is_pack = True if "title" in guess and "year" in matches: if video.series in guess['title']: matches.add("series") # movie else: guess = guessit(self.release_info, {'type': 'movie'}) matches |= guess_matches(video, guess) if video.release_group and "release_group" not in matches and "release_group" in guess: if sanitize_release_group(video.release_group) in sanitize_release_group(guess["release_group"]): matches.add("release_group") self.matches = matches return matches def get_download_link(self, session): return APISubtitle.get_zipped_url(self.page_link, session) def get_video_filename(video): return os.path.splitext(os.path.basename(video.original_name))[0] class SubsceneProvider(Provider, ProviderSubtitleArchiveMixin): """ This currently only searches for the filename on SubScene. It doesn't open every found subtitle page to avoid massive hammering, thus it can't determine whether a subtitle is only-foreign or not. """ subtitle_class = SubsceneSubtitle languages = supported_languages languages.update(set(Language.rebuild(l, forced=True) for l in languages)) languages.update(set(Language.rebuild(l, hi=True) for l in languages)) session = None skip_wrong_fps = False hearing_impaired_verifiable = True only_foreign = False username = None password = None search_throttle = 8 # seconds def __init__(self, only_foreign=False, username=None, password=None): if not all((username, password)): raise ConfigurationError('Username and password must be specified') self.only_foreign = only_foreign self.username = username self.password = password def initialize(self): logger.info("Creating session") self.session = RetryingCFSession() prev_cookies = region.get("subscene_cookies2") if prev_cookies != NO_VALUE: logger.debug("Re-using old subscene cookies: %r", prev_cookies) self.session.cookies.update(prev_cookies) else: logger.debug("Logging in") self.login() def login(self): r = self.session.get("https://subscene.com/account/login") if "Server Error" in r.text: logger.error("Login unavailable; Maintenance?") raise ServiceUnavailable("Login unavailable; Maintenance?") match = re.search(r"", r.text) if match: h = html data = json.loads(h.unescape(match.group(1))) login_url = parse.urljoin(data["siteUrl"], data["loginUrl"]) time.sleep(1.0) r = self.session.post(login_url, { "username": self.username, "password": self.password, data["antiForgery"]["name"]: data["antiForgery"]["value"] }) pep_content = re.search(r"