From b59623ef434cb7868faa4b5e6e0dd2a56449f369 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sat, 12 Dec 2015 00:38:58 +0600 Subject: [PATCH] [funimation] Use mobile webpage for workaround hulu error --- youtube_dl/extractor/funimation.py | 153 +++++++++++++++++------------ 1 file changed, 89 insertions(+), 64 deletions(-) diff --git a/youtube_dl/extractor/funimation.py b/youtube_dl/extractor/funimation.py index 4f382ea6f..8af2e907c 100644 --- a/youtube_dl/extractor/funimation.py +++ b/youtube_dl/extractor/funimation.py @@ -18,7 +18,7 @@ from ..utils import ( class FunimationIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?funimation\.com/shows/[^/]+/videos/official/(?P[^/?#&]+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.funimation.com/shows/air/videos/official/breeze', 'info_dict': { 'id': '658', @@ -28,19 +28,17 @@ class FunimationIE(InfoExtractor): 'description': 'md5:1769f43cd5fc130ace8fd87232207892', 'thumbnail': 're:https?://.*\.jpg', }, - } - - def _download_webpage(self, url_or_request, video_id, note='Downloading webpage'): - HEADERS = { - 'User-Agent': 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0', - } - if isinstance(url_or_request, compat_urllib_request.Request): - for header, value in HEADERS.items(): - url_or_request.add_header(header, value) - else: - url_or_request = sanitized_Request(url_or_request, headers=HEADERS) - response = super(FunimationIE, self)._download_webpage(url_or_request, video_id, note) - return response + }, { + 'url': 'http://www.funimation.com/shows/hacksign/videos/official/role-play', + 'info_dict': { + 'id': '31128', + 'display_id': 'role-play', + 'ext': 'mp4', + 'title': '.hack//SIGN - 1 - Role Play', + 'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd', + 'thumbnail': 're:https?://.*\.jpg', + }, + }] def _login(self): (username, password) = self._get_login_info() @@ -51,6 +49,7 @@ class FunimationIE(InfoExtractor): 'password_field': password, })) login_request = sanitized_Request('http://www.funimation.com/login', data, headers={ + 'User-Agent': 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0', 'Content-Type': 'application/x-www-form-urlencoded' }) login = self._download_webpage( @@ -64,15 +63,8 @@ class FunimationIE(InfoExtractor): def _real_extract(self, url): display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) - - items = self._parse_json( - self._search_regex( - r'var\s+playersData\s*=\s*(\[.+?\]);\n', - webpage, 'players data'), - display_id)[0]['playlist'][0]['items'] - - item = next(item for item in items if item.get('itemAK') == display_id) + errors = [] + formats = [] ERRORS_MAP = { 'ERROR_MATURE_CONTENT_LOGGED_IN': 'matureContentLoggedIn', @@ -87,48 +79,79 @@ class FunimationIE(InfoExtractor): 'ERROR_STREAM_NOT_FOUND': 'streamNotFound', } - error_messages = {} - video_error_messages = self._search_regex( - r'var\s+videoErrorMessages\s*=\s*({.+?});\n', - webpage, 'error messages', default=None) - if video_error_messages: - error_messages_json = self._parse_json(video_error_messages, display_id, fatal=False) - if error_messages_json: - for _, error in error_messages_json.items(): - type_ = error.get('type') - description = error.get('description') - content = error.get('content') - if type_ == 'text' and description and content: - error_message = ERRORS_MAP.get(description) - if error_message: - error_messages[error_message] = content - - errors = [] - formats = [] - for video in item['videoSet']: - auth_token = video.get('authToken') - if not auth_token: - continue - funimation_id = video.get('FUNImationID') or video.get('videoId') - if not auth_token.startswith('?'): - auth_token = '?%s' % auth_token - for quality in ('sd', 'hd', 'hd1080'): - format_url = video.get('%sUrl' % quality) - if not format_url: + USER_AGENTS = ( + # PC UA is served with m3u8 that provides some bonus lower quality formats + ('pc', 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0'), + # Mobile UA allows to extract direct links and also does not fail when + # PC UA fails with hulu error (e.g. + # http://www.funimation.com/shows/hacksign/videos/official/role-play) + ('mobile', 'Mozilla/5.0 (Linux; Android 4.4.2; Nexus 4 Build/KOT49H) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.114 Mobile Safari/537.36'), + ) + + for kind, user_agent in USER_AGENTS: + request = sanitized_Request(url) + request.add_header('User-Agent', user_agent) + webpage = self._download_webpage( + request, display_id, 'Downloading %s webpage' % kind) + + items = self._parse_json( + self._search_regex( + r'var\s+playersData\s*=\s*(\[.+?\]);\n', + webpage, 'players data'), + display_id)[0]['playlist'][0]['items'] + + item = next(item for item in items if item.get('itemAK') == display_id) + + error_messages = {} + video_error_messages = self._search_regex( + r'var\s+videoErrorMessages\s*=\s*({.+?});\n', + webpage, 'error messages', default=None) + if video_error_messages: + error_messages_json = self._parse_json(video_error_messages, display_id, fatal=False) + if error_messages_json: + for _, error in error_messages_json.items(): + type_ = error.get('type') + description = error.get('description') + content = error.get('content') + if type_ == 'text' and description and content: + error_message = ERRORS_MAP.get(description) + if error_message: + error_messages[error_message] = content + + for video in item.get('videoSet', []): + auth_token = video.get('authToken') + if not auth_token: continue - if not format_url.startswith(('http', '//')): - errors.append(format_url) - continue - if determine_ext(format_url) == 'm3u8': - m3u8_formats = self._extract_m3u8_formats( - format_url + auth_token, display_id, 'mp4', entry_protocol='m3u8_native', - m3u8_id=funimation_id or 'hls', fatal=False) - if m3u8_formats: - formats.extend(m3u8_formats) - else: - formats.append({ - 'url': format_url + auth_token, - }) + funimation_id = video.get('FUNImationID') or video.get('videoId') + preference = 1 if video.get('languageMode') == 'dub' else 0 + if not auth_token.startswith('?'): + auth_token = '?%s' % auth_token + for quality in ('sd', 'hd', 'hd1080'): + format_url = video.get('%sUrl' % quality) + if not format_url: + continue + if not format_url.startswith(('http', '//')): + errors.append(format_url) + continue + if determine_ext(format_url) == 'm3u8': + m3u8_formats = self._extract_m3u8_formats( + format_url + auth_token, display_id, 'mp4', entry_protocol='m3u8_native', + preference=preference, m3u8_id=funimation_id or 'hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + else: + f = { + 'url': format_url + auth_token, + 'format_id': funimation_id, + 'preference': preference, + } + mobj = re.search(r'(?P\d+)-(?P\d+)[Kk]', format_url) + if mobj: + f.update({ + 'height': int(mobj.group('height')), + 'tbr': int(mobj.group('tbr')), + }) + formats.append(f) if not formats and errors: raise ExtractorError( @@ -136,6 +159,8 @@ class FunimationIE(InfoExtractor): % (self.IE_NAME, clean_html(error_messages.get(errors[0], errors[0]))), expected=True) + self._sort_formats(formats) + title = item['title'] artist = item.get('artist') if artist: