def extract_subpage(q, name, torrent, size, seeds, peers, info_hash, referer): try: log.debug("[%s] Getting subpage at %s" % (provider, repr(torrent))) except Exception as e: import traceback log.error("[%s] Subpage logging failed with: %s" % (provider, repr(e))) map(log.debug, traceback.format_exc().split("\n")) # New client instance, otherwise it's race conditions all over the place subclient = Client() subclient.passkey = client.passkey headers = {} if "subpage_mode" in definition: if definition["subpage_mode"] == "xhr": headers['X-Requested-With'] = 'XMLHttpRequest' headers['Content-Language'] = '' if referer: headers['Referer'] = referer uri = torrent.split('|') # Split cookies for private trackers subclient.open(uri[0].encode('utf-8'), headers=headers) if 'bittorrent' in subclient.headers.get('content-type', ''): log.debug('[%s] bittorrent content-type for %s' % (provider, repr(torrent))) if len(uri) > 1: # Stick back cookies if needed torrent = '%s|%s' % (torrent, uri[1]) else: try: torrent = extract_from_page(provider, subclient.content) if torrent and not torrent.startswith('magnet') and len( uri) > 1: # Stick back cookies if needed torrent = '%s|%s' % (torrent, uri[1]) except Exception as e: import traceback log.error( "[%s] Subpage extraction for %s failed with: %s" % (provider, repr(uri[0]), repr(e))) map(log.debug, traceback.format_exc().split("\n")) log.debug("[%s] Subpage torrent for %s: %s" % (provider, repr(uri[0]), torrent)) ret = (name, info_hash, torrent, size, seeds, peers) q.put_nowait(ret)
def load_providers(path, custom=False): """ Definitions loader for json files Args: path (str): Path to json file to be loaded custom (bool): Boolean flag to specify if this is a custom provider """ if not os.path.exists(path): return try: with open(path) as file: providers = json.load(file) for provider in providers: update_definitions(provider, providers[provider], custom) except Exception as e: import traceback log.error("Failed importing providers from %s: %s", path, repr(e)) map(log.error, traceback.format_exc().split("\n"))
def load_overrides(path, custom=False): """ Overrides loader for Python files Note: Overrides must be in an ``overrides`` dictionary. Args: path (str): Path to Python file to be loaded custom (bool): Boolean flag to specify if this is a custom overrides file """ try: if custom: sys.path.append(path) from overrides import overrides log.debug("Imported overrides: %s", repr(overrides)) for provider in overrides: update_definitions(provider, overrides[provider]) log.info("Successfully loaded overrides from %s", os.path.join(path, "overrides.py")) except Exception as e: import traceback log.error("Failed importing %soverrides: %s", "custom " if custom else "", repr(e)) map(log.error, traceback.format_exc().split("\n"))
def search(payload, method="general"): """ Main search entrypoint Args: payload (dict): Search payload from projectx. method (str): Type of search, can be ``general``, ``movie``, ``show``, ``season`` or ``anime`` Returns: list: All filtered results in the format projectx expects """ log.debug("Searching with payload (%s): %s" % (method, repr(payload))) if method == 'general': if 'query' in payload: payload['title'] = payload['query'] payload['titles'] = {'source': payload['query']} else: payload = { 'title': payload, 'titles': { 'source': payload }, } payload['titles'] = dict( (k.lower(), v) for k, v in payload['titles'].iteritems()) # If titles[] exists in payload and there are special chars in titles[source] # then we set a flag to possibly modify the search query payload['has_special'] = 'titles' in payload and \ bool(payload['titles']) and \ 'source' in payload['titles'] and \ any(c in payload['titles']['source'] for c in special_chars) if payload['has_special']: log.debug( "Query title contains special chars, so removing any quotes in the search query" ) if 'proxy_url' not in payload: payload['proxy_url'] = '' if 'internal_proxy_url' not in payload: payload['internal_proxy_url'] = '' if 'projectx_url' not in payload: payload['projectx_url'] = '' if 'silent' not in payload: payload['silent'] = False if 'skip_auth' not in payload: payload['skip_auth'] = False global request_time global provider_names global provider_results global available_providers provider_names = [] provider_results = [] available_providers = 0 request_time = time.time() providers = get_enabled_providers(method) if len(providers) == 0: if not payload['silent']: notify(translation(32060), image=get_icon_path()) log.error("No providers enabled") return [] log.info( "Burstin' with %s" % ", ".join([definitions[provider]['name'] for provider in providers])) if get_setting('kodi_language', bool): kodi_language = xbmc.getLanguage(xbmc.ISO_639_1) if not kodi_language: log.warning("Kodi returned empty language code...") elif 'titles' not in payload or not payload['titles']: log.info("No translations available...") elif payload['titles'] and kodi_language not in payload['titles']: log.info("No '%s' translation available..." % kodi_language) p_dialog = xbmcgui.DialogProgressBG() if not payload['silent']: p_dialog.create('projectx [COLOR FFFF6B00]Burst[/COLOR]', translation(32061)) for provider in providers: available_providers += 1 provider_names.append(definitions[provider]['name']) task = Thread(target=run_provider, args=(provider, payload, method)) task.start() providers_time = time.time() total = float(available_providers) # Exit if all providers have returned results or timeout reached, check every 100ms while time.time() - providers_time < timeout and available_providers > 0: timer = time.time() - providers_time log.debug("Timer: %ds / %ds" % (timer, timeout)) if timer > timeout: break message = translation( 32062 ) % available_providers if available_providers > 1 else translation( 32063) if not payload['silent']: p_dialog.update(int((total - available_providers) / total * 100), message=message) time.sleep(0.25) if not payload['silent']: p_dialog.close() del p_dialog if available_providers > 0: message = u', '.join(provider_names) message = message + translation(32064) log.warning(message.encode('utf-8')) if not payload['silent']: notify(message, ADDON_ICON) log.debug("all provider_results: %s" % repr(provider_results)) filtered_results = apply_filters(provider_results) log.debug("all filtered_results: %s" % repr(filtered_results)) log.info("Providers returned %d results in %s seconds" % (len(filtered_results), round(time.time() - request_time, 2))) return filtered_results
def extract_torrents(provider, client): """ Main torrent extraction generator for non-API based providers Args: provider (str): Provider ID client (Client): Client class instance Yields: tuple: A torrent result """ definition = definitions[provider] definition = get_alias(definition, get_setting("%s_alias" % provider)) log.debug("[%s] Extracting torrents from %s using definitions: %s" % (provider, provider, repr(definition))) if not client.content: if get_setting("use_debug_parser", bool): log.debug("[%s] Parser debug | Page content is empty" % provider) raise StopIteration dom = Html().feed(client.content) key_search = get_search_query(definition, "key") row_search = get_search_query(definition, "row") name_search = get_search_query(definition, "name") torrent_search = get_search_query(definition, "torrent") info_hash_search = get_search_query(definition, "infohash") size_search = get_search_query(definition, "size") seeds_search = get_search_query(definition, "seeds") peers_search = get_search_query(definition, "peers") referer_search = get_search_query(definition, "referer") log.debug("[%s] Parser: %s" % (provider, repr(definition['parser']))) q = Queue() threads = [] needs_subpage = 'subpage' in definition and definition['subpage'] if needs_subpage: def extract_subpage(q, name, torrent, size, seeds, peers, info_hash, referer): try: log.debug("[%s] Getting subpage at %s" % (provider, repr(torrent))) except Exception as e: import traceback log.error("[%s] Subpage logging failed with: %s" % (provider, repr(e))) map(log.debug, traceback.format_exc().split("\n")) # New client instance, otherwise it's race conditions all over the place subclient = Client() subclient.passkey = client.passkey headers = {} if "subpage_mode" in definition: if definition["subpage_mode"] == "xhr": headers['X-Requested-With'] = 'XMLHttpRequest' headers['Content-Language'] = '' if referer: headers['Referer'] = referer uri = torrent.split('|') # Split cookies for private trackers subclient.open(uri[0].encode('utf-8'), headers=headers) if 'bittorrent' in subclient.headers.get('content-type', ''): log.debug('[%s] bittorrent content-type for %s' % (provider, repr(torrent))) if len(uri) > 1: # Stick back cookies if needed torrent = '%s|%s' % (torrent, uri[1]) else: try: torrent = extract_from_page(provider, subclient.content) if torrent and not torrent.startswith('magnet') and len( uri) > 1: # Stick back cookies if needed torrent = '%s|%s' % (torrent, uri[1]) except Exception as e: import traceback log.error( "[%s] Subpage extraction for %s failed with: %s" % (provider, repr(uri[0]), repr(e))) map(log.debug, traceback.format_exc().split("\n")) log.debug("[%s] Subpage torrent for %s: %s" % (provider, repr(uri[0]), torrent)) ret = (name, info_hash, torrent, size, seeds, peers) q.put_nowait(ret) if not dom: if get_setting("use_debug_parser", bool): log.debug( "[%s] Parser debug | Could not parse DOM from page content" % provider) raise StopIteration if get_setting("use_debug_parser", bool): log.debug( "[%s] Parser debug | Page content: %s" % (provider, client.content.replace('\r', '').replace('\n', ''))) key = eval(key_search) if key_search else "" if key_search and get_setting("use_debug_parser", bool): key_str = key.__str__() log.debug( "[%s] Parser debug | Matched '%s' iteration for query '%s': %s" % (provider, 'key', key_search, key_str.replace('\r', '').replace( '\n', ''))) items = eval(row_search) if get_setting("use_debug_parser", bool): log.debug("[%s] Parser debug | Matched %d items for '%s' query '%s'" % (provider, len(items), 'row', row_search)) for item in items: if get_setting("use_debug_parser", bool): item_str = item.__str__() log.debug( "[%s] Parser debug | Matched '%s' iteration for query '%s': %s" % (provider, 'row', row_search, item_str.replace( '\r', '').replace('\n', ''))) if not item: continue try: name = eval(name_search) if name_search else "" torrent = eval(torrent_search) if torrent_search else "" size = eval(size_search) if size_search else "" seeds = eval(seeds_search) if seeds_search else "" peers = eval(peers_search) if peers_search else "" info_hash = eval(info_hash_search) if info_hash_search else "" referer = eval(referer_search) if referer_search else "" if 'magnet:?' in torrent: torrent = torrent[torrent.find('magnet:?'):] if get_setting("use_debug_parser", bool): log.debug( "[%s] Parser debug | Matched '%s' iteration for query '%s': %s" % (provider, 'name', name_search, name)) log.debug( "[%s] Parser debug | Matched '%s' iteration for query '%s': %s" % (provider, 'torrent', torrent_search, torrent)) log.debug( "[%s] Parser debug | Matched '%s' iteration for query '%s': %s" % (provider, 'size', size_search, size)) log.debug( "[%s] Parser debug | Matched '%s' iteration for query '%s': %s" % (provider, 'seeds', seeds_search, seeds)) log.debug( "[%s] Parser debug | Matched '%s' iteration for query '%s': %s" % (provider, 'peers', peers_search, peers)) if info_hash_search: log.debug( "[%s] Parser debug | Matched '%s' iteration for query '%s': %s" % (provider, 'info_hash', info_hash_search, info_hash)) if referer_search: log.debug( "[%s] Parser debug | Matched '%s' iteration for query '%s': %s" % (provider, 'info_hash', referer_search, referer)) # Pass client cookies with torrent if private if not torrent.startswith('magnet'): user_agent = USER_AGENT if client.passkey: torrent = torrent.replace('PASSKEY', client.passkey) elif client.token: headers = { 'Authorization': client.token, 'User-Agent': user_agent } log.debug("[%s] Appending headers: %s" % (provider, repr(headers))) torrent = append_headers(torrent, headers) log.debug("[%s] Torrent with headers: %s" % (provider, repr(torrent))) else: parsed_url = urlparse(torrent.split('|')[0]) cookie_domain = '{uri.netloc}'.format(uri=parsed_url) cookie_domain = re.sub('www\d*\.', '', cookie_domain) cookies = [] for cookie in client._cookies: if cookie_domain in cookie.domain: cookies.append(cookie) headers = {} if cookies: headers = {'User-Agent': user_agent} log.debug("[%s] Cookies res: %s / %s" % (provider, repr(headers), repr(client.request_headers))) if client.request_headers: headers.update(client.request_headers) if client.url: headers['Referer'] = client.url headers['Origin'] = client.url # Need to set Cookie afterwards to avoid rewriting it with session Cookies headers['Cookie'] = ";".join( ["%s=%s" % (c.name, c.value) for c in cookies]) else: headers = {'User-Agent': user_agent} torrent = append_headers(torrent, headers) if name and torrent and needs_subpage and not torrent.startswith( 'magnet'): if not torrent.startswith('http'): torrent = definition['root_url'] + torrent.encode('utf-8') t = Thread(target=extract_subpage, args=(q, name, torrent, size, seeds, peers, info_hash, referer)) threads.append(t) else: yield (name, info_hash, torrent, size, seeds, peers) except Exception as e: log.error("[%s] Got an exception while parsing results: %s" % (provider, repr(e))) if needs_subpage: log.debug("[%s] Starting subpage threads..." % provider) for t in threads: t.start() for t in threads: t.join() for i in range(q.qsize()): ret = q.get_nowait() log.debug("[%s] Queue %d got: %s" % (provider, i, repr(ret))) yield ret
def open(self, url, language='en', post_data=None, get_data=None, headers=None): """ Opens a connection to a webpage and saves its HTML content in ``self.content`` Args: url (str): The URL to open language (str): The language code for the ``Content-Language`` header post_data (dict): POST data for the request get_data (dict): GET data for the request """ if get_data: url += '?' + urlencode(get_data) log.debug("Opening URL: %s" % repr(url)) if self.session.proxies: log.debug("Proxies: %s" % (repr(self.session.proxies))) self._read_cookies(url) self.session.cookies = self._cookies # log.debug("Cookies for %s: %s" % (repr(url), repr(self._cookies))) # Default headers for any request. Pretend like we are the usual browser. req_headers = { 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9', 'Accept-Language': 'en-EN,en;q=0.9,en-US;q=0.8,en;q=0.7,uk;q=0.6,pl;q=0.5', 'Cache-Control': 'no-cache', 'Content-Language': language, 'Origin': url, 'Referer': url, 'User-Agent': self.user_agent } # If headers passed to open() call - we overwrite headers. if headers: for key, value in headers.iteritems(): if key == ':path': u = urlparse(url) value = u.path if value: req_headers[key] = value elif key.capitalize() in req_headers: del req_headers[key.capitalize()] if self.token: req_headers["Authorization"] = self.token req = None if post_data: req = requests.Request('POST', url, data=post_data, headers=req_headers) else: req = requests.Request('GET', url, headers=req_headers) prepped = self.session.prepare_request(req) self.request_headers = prepped.headers try: self._good_spider() with self.session.send(prepped) as response: self.headers = response.headers self.status = response.status_code self.url = response.url self._save_cookies() if self.response_charset: self.content = response.content.decode( self.response_charset, 'ignore') else: self.content = response.text except requests.exceptions.InvalidSchema as e: # If link points to a magnet: then it can be used as a content matches = re.findall( 'No connection adapters were found for \'(.*?)\'', str(e)) if matches: self.content = matches[0] return True import traceback log.error("%s failed with %s:" % (repr(url), repr(e))) map(log.debug, traceback.format_exc().split("\n")) except Exception as e: import traceback log.error("%s failed with %s:" % (repr(url), repr(e))) map(log.debug, traceback.format_exc().split("\n")) log.debug("Status for %s : %s" % (repr(url), str(self.status))) return self.status == 200
def process(provider, generator, filtering, has_special, verify_name=True, verify_size=True, skip_auth=False): """ Method for processing provider results using its generator and Filtering class instance Args: provider (str): Provider ID generator (function): Generator method, can be either ``extract_torrents`` or ``extract_from_api`` filtering (Filtering): Filtering class instance has_special (bool): Whether title contains special chars verify_name (bool): Whether to double-check the results' names match the query or not verify_size (bool): Whether to check the results' file sizes """ log.debug("[%s] execute_process for %s with %s" % (provider, provider, repr(generator))) definition = definitions[provider] definition = get_alias(definition, get_setting("%s_alias" % provider)) client = Client(info=filtering.info, request_charset=definition['charset'], response_charset=definition['response_charset']) token = None logged_in = False token_auth = False if get_setting('kodi_language', bool): kodi_language = xbmc.getLanguage(xbmc.ISO_639_1) if kodi_language: filtering.kodi_language = kodi_language language_exceptions = get_setting('language_exceptions') if language_exceptions.strip().lower(): filtering.language_exceptions = re.split(r',\s?', language_exceptions) log.debug("[%s] Queries: %s" % (provider, filtering.queries)) log.debug("[%s] Extras: %s" % (provider, filtering.extras)) for query, extra in zip(filtering.queries, filtering.extras): log.debug("[%s] Before keywords - Query: %s - Extra: %s" % (provider, repr(query), repr(extra))) if has_special: # Removing quotes, surrounding {title*} keywords, when title contains special chars query = re.sub("[\"']({title.*?})[\"']", '\\1', query) query = filtering.process_keywords(provider, query) extra = filtering.process_keywords(provider, extra) if extra == '-' and filtering.results: continue try: if 'charset' in definition and definition['charset'] and 'utf' not in definition['charset'].lower(): query = urllib.quote(query.encode(definition['charset'])) extra = urllib.quote(extra.encode(definition['charset'])) else: query = urllib.quote(query.encode('utf-8')) extra = urllib.quote(extra.encode('utf-8')) except Exception as e: log.debug("[%s] Could not quote the query (%s): %s" % (provider, query, e)) pass log.debug("[%s] After keywords - Query: %s - Extra: %s" % (provider, repr(query), repr(extra))) if not query: return filtering.results url_search = filtering.url.replace('QUERY', query) if extra and extra != '-': url_search = url_search.replace('EXTRA', extra) else: url_search = url_search.replace('EXTRA', '') url_search = url_search.replace(' ', definition['separator']) if definition['separator'] != '%20': url_search = url_search.replace('%20', definition['separator']) # MagnetDL fix... url_search = url_search.replace('FIRSTLETTER', query[:1]) # Creating the payload for POST method if 'post_data' in definition and not filtering.post_data: filtering.post_data = eval(definition['post_data']) payload = dict() for key, value in filtering.post_data.iteritems(): if 'QUERY' in value: payload[key] = filtering.post_data[key].replace('QUERY', query) else: payload[key] = filtering.post_data[key] payload[key] = urllib.unquote(payload[key]) # Creating the payload for GET method headers = None data = None if filtering.get_data: data = dict() for key, value in filtering.get_data.iteritems(): if 'QUERY' in value: data[key] = filtering.get_data[key].replace('QUERY', query) else: data[key] = filtering.get_data[key] log.debug("- %s query: %s" % (provider, repr(query))) log.debug("-- %s url_search before token: %s" % (provider, repr(url_search))) log.debug("--- %s using POST payload: %s" % (provider, repr(payload))) log.debug("----%s filtering with post_data: %s" % (provider, repr(filtering.post_data))) # Set search's "title" in filtering to double-check results' names if 'filter_title' in definition and definition['filter_title']: filtering.filter_title = True filtering.title = query if 'initial_url' in definition and definition['initial_url']: url = definition['initial_url'] if not url.startswith('http'): url = definition['root_url'] + url client.open(url) if token: log.info('[%s] Reusing existing token' % provider) url_search = url_search.replace('TOKEN', token) elif 'token' in definition: token_url = definition['base_url'] + definition['token'] log.debug("[%s] Getting token for %s at %s" % (provider, provider, repr(token_url))) client.open(token_url.encode('utf-8')) try: token_data = json.loads(client.content) except: log.error('%s: Failed to get token for %s' % (provider, repr(url_search))) return filtering.results log.debug("[%s] Token response for %s: %s" % (provider, provider, repr(token_data))) if 'token' in token_data: token = token_data['token'] log.debug("[%s] Got token for %s: %s" % (provider, provider, repr(token))) url_search = url_search.replace('TOKEN', token) else: log.warning('%s: Unable to get token for %s' % (provider, repr(url_search))) if logged_in: log.info("[%s] Reusing previous login" % provider) elif token_auth: log.info("[%s] Reusing previous token authorization" % provider) elif 'private' in definition and definition['private']: username = get_setting('%s_username' % provider, unicode) password = get_setting('%s_password' % provider, unicode) passkey = get_setting('%s_passkey' % provider, unicode) if not username and not password and not passkey: for addon_name in ('script.magnetic.%s' % provider, 'script.magnetic.%s-mc' % provider): for setting in ('username', 'password'): try: value = xbmcaddon.Addon(addon_name).getSetting(setting) set_setting('%s_%s' % (provider, setting), value) if setting == 'username': username = value if setting == 'password': password = value except: pass if username: client.username = username url_search = url_search.replace('USERNAME', username) if passkey: logged_in = True client.passkey = passkey url_search = url_search.replace('PASSKEY', passkey) elif 'login_object' in definition and definition['login_object']: login_object = None login_headers = None logged_in = skip_auth try: login_object = definition['login_object'].replace('USERNAME', 'u"%s"' % username).replace('PASSWORD', 'u"%s"' % password) except Exception as e: log.error("Could not make login object for %s: %s" % (provider, e)) try: if 'login_headers' in definition and definition['login_headers']: login_headers = eval(definition['login_headers']) except Exception as e: log.error("Could not make login headers for %s: %s" % (provider, e)) # TODO generic flags in definitions for those... if 'csrf_token' in definition and definition['csrf_token']: client.open(definition['root_url'] + definition['login_path']) if client.content: csrf_token = re.search(r'name=\"_?csrf_token\" value=\"(.*?)\"', client.content) if csrf_token: login_object = login_object.replace('CSRF_TOKEN', '"%s"' % csrf_token.group(1)) else: logged_in = True if 'token_auth' in definition: # log.debug("[%s] logging in with: %s" % (provider, login_object)) if client.open(definition['root_url'] + definition['token_auth'], post_data=eval(login_object)): try: token_data = json.loads(client.content) except: log.error('%s: Failed to get token from %s' % (provider, definition['token_auth'])) return filtering.results log.debug("[%s] Token response for %s: %s" % (provider, provider, repr(token_data))) if 'token' in token_data: client.token = token_data['token'] log.debug("[%s] Auth token for %s: %s" % (provider, provider, repr(client.token))) else: log.error('[%s] Unable to get auth token for %s' % (provider, repr(url_search))) return filtering.results log.info('[%s] Token auth successful' % provider) token_auth = True else: log.error("[%s] Token auth failed with response: %s" % (provider, repr(client.content))) return filtering.results elif not logged_in and client.login(definition['root_url'], definition['login_path'], eval(login_object), login_headers, definition['login_failed']): log.info('[%s] Login successful' % provider) logged_in = True elif not logged_in: log.error("[%s] Login failed: %s", provider, client.status) log.debug("[%s] Failed login content: %s", provider, repr(client.content)) return filtering.results if logged_in: if provider == 'hd-torrents': client.open(definition['root_url'] + '/torrents.php') csrf_token = re.search(r'name="csrfToken" value="(.*?)"', client.content) url_search = url_search.replace("CSRF_TOKEN", csrf_token.group(1)) log.info("[%s] > %s search URL: %s" % (provider, definition['name'].rjust(longest), url_search)) if 'headers' in definition and definition['headers']: headers = eval(definition['headers']) log.info("[%s] > %s headers: %s" % (provider, definition['name'].rjust(longest), headers)) client.open(url_search.encode('utf-8'), post_data=payload, get_data=data, headers=headers) filtering.results.extend( generate_payload(provider, generator(provider, client), filtering, verify_name, verify_size)) return filtering.results
return d # Load providers load_providers(os.path.join(ADDON_PATH, 'burst', 'providers', 'providers.json')) # Load providers overrides load_overrides(os.path.join(ADDON_PATH, 'burst', 'providers')) # Load user's custom providers custom_providers = os.path.join(xbmc.translatePath(ADDON_PROFILE), "providers") if not os.path.exists(custom_providers): try: os.makedirs(custom_providers) except Exception as e: log.error("Unable to create custom providers folder: %s", repr(e)) pass for provider_file in glob(os.path.join(custom_providers, "*.json")): log.info("Importing and enabling %s" % provider_file) load_providers(provider_file, custom=True) # Load user's custom overrides custom_overrides = xbmc.translatePath(ADDON_PROFILE) if os.path.exists(os.path.join(custom_overrides, 'overrides.py')): load_overrides(custom_overrides, custom=True) # Load json overrides load_providers(os.path.join(xbmc.translatePath(ADDON_PROFILE), 'overrides.json')) # Setting mandatory fields to their default values for each provider. for provider in definitions: