X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9f2b33881274af98a9145c533a1d295fad71521a..d9bd2d488401884d00ee455f2a161a5011f7457d:/youtube_dl/extractor/discovery.py?ds=inline diff --git a/youtube_dl/extractor/discovery.py b/youtube_dl/extractor/discovery.py index ce680a9..3589bd4 100644 --- a/youtube_dl/extractor/discovery.py +++ b/youtube_dl/extractor/discovery.py @@ -1,15 +1,23 @@ from __future__ import unicode_literals -from .common import InfoExtractor +import random +import re +import string + +from .discoverygo import DiscoveryGoBaseIE +from ..compat import ( + compat_str, + compat_urllib_parse_unquote, +) from ..utils import ( - parse_duration, - parse_iso8601, + ExtractorError, + try_get, ) -from ..compat import compat_str +from ..compat import compat_HTTPError -class DiscoveryIE(InfoExtractor): - _VALID_URL = r'''(?x)http://(?:www\.)?(?: +class DiscoveryIE(DiscoveryGoBaseIE): + _VALID_URL = r'''(?x)https?://(?:www\.)?(?P discovery| investigationdiscovery| discoverylife| @@ -19,62 +27,76 @@ class DiscoveryIE(InfoExtractor): sciencechannel| tlc| velocity - )\.com/(?:[^/]+/)*(?P[^./?#]+)''' + )\.com(?P/tv-shows/[^/]+/(?:video|full-episode)s/(?P[^./?#]+))''' _TESTS = [{ - 'url': 'http://www.discovery.com/tv-shows/mythbusters/videos/mission-impossible-outtakes.htm', + 'url': 'https://www.discovery.com/tv-shows/cash-cab/videos/dave-foley', 'info_dict': { - 'id': '20769', + 'id': '5a2d9b4d6b66d17a5026e1fd', 'ext': 'mp4', - 'title': 'Mission Impossible Outtakes', - 'description': ('Watch Jamie Hyneman and Adam Savage practice being' - ' each other -- to the point of confusing Jamie\'s dog -- and ' - 'don\'t miss Adam moon-walking as Jamie ... behind Jamie\'s' - ' back.'), - 'duration': 156, - 'timestamp': 1302032462, - 'upload_date': '20110405', + 'title': 'Dave Foley', + 'description': 'md5:4b39bcafccf9167ca42810eb5f28b01f', + 'duration': 608, }, 'params': { 'skip_download': True, # requires ffmpeg } }, { - 'url': 'http://www.discovery.com/tv-shows/mythbusters/videos/mythbusters-the-simpsons', - 'info_dict': { - 'id': 'mythbusters-the-simpsons', - 'title': 'MythBusters: The Simpsons', - }, - 'playlist_mincount': 10, - }, { - 'url': 'http://www.animalplanet.com/longfin-eels-maneaters/', - 'info_dict': { - 'id': '78326', - 'ext': 'mp4', - 'title': 'Longfin Eels: Maneaters?', - 'description': 'Jeremy Wade tests whether or not New Zealand\'s longfin eels are man-eaters by covering himself in fish guts and getting in the water with them.', - 'upload_date': '20140725', - 'timestamp': 1406246400, - 'duration': 116, - }, + 'url': 'https://www.investigationdiscovery.com/tv-shows/final-vision/full-episodes/final-vision', + 'only_matching': True, }] + _GEO_COUNTRIES = ['US'] + _GEO_BYPASS = False def _real_extract(self, url): - display_id = self._match_id(url) - info = self._download_json(url + '?flat=1', display_id) + site, path, display_id = re.match(self._VALID_URL, url).groups() + webpage = self._download_webpage(url, display_id) + + react_data = self._parse_json(self._search_regex( + r'window\.__reactTransmitPacket\s*=\s*({.+?});', + webpage, 'react data'), display_id) + content_blocks = react_data['layout'][path]['contentBlocks'] + video = next(cb for cb in content_blocks if cb.get('type') == 'video')['content']['items'][0] + video_id = video['id'] + + access_token = None + cookies = self._get_cookies(url) + + # prefer Affiliate Auth Token over Anonymous Auth Token + auth_storage_cookie = cookies.get('eosAf') or cookies.get('eosAn') + if auth_storage_cookie and auth_storage_cookie.value: + auth_storage = self._parse_json(compat_urllib_parse_unquote( + compat_urllib_parse_unquote(auth_storage_cookie.value)), + video_id, fatal=False) or {} + access_token = auth_storage.get('a') or auth_storage.get('access_token') - video_title = info.get('playlist_title') or info.get('video_title') + if not access_token: + access_token = self._download_json( + 'https://www.%s.com/anonymous' % site, display_id, query={ + 'authRel': 'authorization', + 'client_id': try_get( + react_data, lambda x: x['application']['apiClientId'], + compat_str) or '3020a40c2356a645b4b4', + 'nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]), + 'redirectUri': 'https://fusion.ddmcdn.com/app/mercury-sdk/180/redirectHandler.html?https://www.%s.com' % site, + })['access_token'] - entries = [{ - 'id': compat_str(video_info['id']), - 'formats': self._extract_m3u8_formats( - video_info['src'], display_id, 'mp4', 'm3u8_native', m3u8_id='hls', - note='Download m3u8 information for video %d' % (idx + 1)), - 'title': video_info['title'], - 'description': video_info.get('description'), - 'duration': parse_duration(video_info.get('video_length')), - 'webpage_url': video_info.get('href') or video_info.get('url'), - 'thumbnail': video_info.get('thumbnailURL'), - 'alt_title': video_info.get('secondary_title'), - 'timestamp': parse_iso8601(video_info.get('publishedDate')), - } for idx, video_info in enumerate(info['playlist'])] + try: + stream = self._download_json( + 'https://api.discovery.com/v1/streaming/video/' + video_id, + display_id, headers={ + 'Authorization': 'Bearer ' + access_token, + }) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 403): + e_description = self._parse_json( + e.cause.read().decode(), display_id)['description'] + if 'resource not available for country' in e_description: + self.raise_geo_restricted(countries=self._GEO_COUNTRIES) + if 'Authorized Networks' in e_description: + raise ExtractorError( + 'This video is only available via cable service provider subscription that' + ' is not currently supported. You may want to use --cookies.', expected=True) + raise ExtractorError(e_description) + raise - return self.playlist_result(entries, display_id, video_title) + return self._extract_video_info(video, stream, display_id)