X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/7ceb2ec430c3363e0140a0519402428f36dc472e..5a9d00a5190582368d533f12db7638e735d7f7c8:/youtube_dl/extractor/vice.py?ds=inline diff --git a/youtube_dl/extractor/vice.py b/youtube_dl/extractor/vice.py index 6b93afa..bcc2869 100644 --- a/youtube_dl/extractor/vice.py +++ b/youtube_dl/extractor/vice.py @@ -1,38 +1,273 @@ +# coding: utf-8 +from __future__ import unicode_literals + import re +import time +import hashlib +import json +from .adobepass import AdobePassIE +from .youtube import YoutubeIE from .common import InfoExtractor -from .ooyala import OoyalaIE -from ..utils import ExtractorError +from ..compat import compat_HTTPError +from ..utils import ( + int_or_none, + parse_age_limit, + str_or_none, + parse_duration, + ExtractorError, + extract_attributes, +) -class ViceIE(InfoExtractor): - _VALID_URL = r'http://www.vice.com/.*?/(?P.+)' +class ViceBaseIE(AdobePassIE): + def _extract_preplay_video(self, url, locale, webpage): + watch_hub_data = extract_attributes(self._search_regex( + r'(?s)()', webpage, 'watch hub')) + video_id = watch_hub_data['vms-id'] + title = watch_hub_data['video-title'] - _TEST = { - u'url': u'http://www.vice.com/Fringes/cowboy-capitalists-part-1', - u'file': u'43cW1mYzpia9IlestBjVpd23Yu3afAfp.mp4', - u'info_dict': { - u'title': u'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov', + query = {} + is_locked = watch_hub_data.get('video-locked') == '1' + if is_locked: + resource = self._get_mvpd_resource( + 'VICELAND', title, video_id, + watch_hub_data.get('video-rating')) + query['tvetoken'] = self._extract_mvpd_auth( + url, video_id, 'VICELAND', resource) + + # signature generation algorithm is reverse engineered from signatureGenerator in + # webpack:///../shared/~/vice-player/dist/js/vice-player.js in + # https://www.viceland.com/assets/common/js/web.vendor.bundle.js + exp = int(time.time()) + 14400 + query.update({ + 'exp': exp, + 'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(), + }) + + try: + host = 'www.viceland' if is_locked else self._PREPLAY_HOST + preplay = self._download_json( + 'https://%s.com/%s/preplay/%s' % (host, locale, video_id), + video_id, query=query) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400: + error = json.loads(e.cause.read().decode()) + raise ExtractorError('%s said: %s' % ( + self.IE_NAME, error['details']), expected=True) + raise + + video_data = preplay['video'] + base = video_data['base'] + uplynk_preplay_url = preplay['preplayURL'] + episode = video_data.get('episode', {}) + channel = video_data.get('channel', {}) + + subtitles = {} + cc_url = preplay.get('ccURL') + if cc_url: + subtitles['en'] = [{ + 'url': cc_url, + }] + + return { + '_type': 'url_transparent', + 'url': uplynk_preplay_url, + 'id': video_id, + 'title': title, + 'description': base.get('body') or base.get('display_body'), + 'thumbnail': watch_hub_data.get('cover-image') or watch_hub_data.get('thumbnail'), + 'duration': int_or_none(video_data.get('video_duration')) or parse_duration(watch_hub_data.get('video-duration')), + 'timestamp': int_or_none(video_data.get('created_at'), 1000), + 'age_limit': parse_age_limit(video_data.get('video_rating')), + 'series': video_data.get('show_title') or watch_hub_data.get('show-title'), + 'episode_number': int_or_none(episode.get('episode_number') or watch_hub_data.get('episode')), + 'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')), + 'season_number': int_or_none(watch_hub_data.get('season')), + 'season_id': str_or_none(episode.get('season_id')), + 'uploader': channel.get('base', {}).get('title') or watch_hub_data.get('channel-title'), + 'uploader_id': str_or_none(channel.get('id')), + 'subtitles': subtitles, + 'ie_key': 'UplynkPreplay', + } + + +class ViceIE(ViceBaseIE): + IE_NAME = 'vice' + _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:(?P[^/]+)/)?videos?/(?P[^/?#&]+)' + + _TESTS = [{ + 'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab', + 'md5': '7d3ae2f9ba5f196cdd9f9efd43657ac2', + 'info_dict': { + 'id': 'N2bzkydjraWDGwnt8jAttCF6Y0PDv4Zj', + 'ext': 'flv', + 'title': 'Monkey Labs of Holland', + 'description': 'md5:92b3c7dcbfe477f772dd4afa496c9149', + }, + 'add_ie': ['Ooyala'], + }, { + 'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56', + 'info_dict': { + 'id': '5816510690b70e6c5fd39a56', + 'ext': 'mp4', + 'uploader': 'Waypoint', + 'title': 'The Signal From Tölva', + 'description': 'md5:3927e3c79f9e8094606a2b3c5b5e55d5', + 'uploader_id': '57f7d621e05ca860fa9ccaf9', + 'timestamp': 1477941983, + 'upload_date': '20161031', + }, + 'params': { + # m3u8 download + 'skip_download': True, }, - u'params': { - # Requires ffmpeg (m3u8 manifest) - u'skip_download': True, + 'add_ie': ['UplynkPreplay'], + }, { + 'url': 'https://video.vice.com/alps/video/ulfs-wien-beruchtigste-grafitti-crew-part-1/581b12b60a0e1f4c0fb6ea2f', + 'info_dict': { + 'id': '581b12b60a0e1f4c0fb6ea2f', + 'ext': 'mp4', + 'title': 'ULFs - Wien berüchtigste Grafitti Crew - Part 1', + 'description': '

Zwischen Hinterzimmer-Tattoos und U-Bahnschächten erzählen uns die Ulfs, wie es ist, "süchtig nach Sachbeschädigung" zu sein.

', + 'uploader': 'VICE', + 'uploader_id': '57a204088cb727dec794c67b', + 'timestamp': 1485368119, + 'upload_date': '20170125', + 'age_limit': 14, }, + 'params': { + # AES-encrypted m3u8 + 'skip_download': True, + }, + 'add_ie': ['UplynkPreplay'], + }, { + 'url': 'https://video.vice.com/en_us/video/pizza-show-trailer/56d8c9a54d286ed92f7f30e4', + 'only_matching': True, + }] + _PREPLAY_HOST = 'video.vice' + + def _real_extract(self, url): + locale, video_id = re.match(self._VALID_URL, url).groups() + webpage, urlh = self._download_webpage_handle(url, video_id) + embed_code = self._search_regex( + r'embedCode=([^&\'"]+)', webpage, + 'ooyala embed code', default=None) + if embed_code: + return self.url_result('ooyala:%s' % embed_code, 'Ooyala') + youtube_id = self._search_regex( + r'data-youtube-id="([^"]+)"', webpage, 'youtube id', default=None) + if youtube_id: + return self.url_result(youtube_id, 'Youtube') + return self._extract_preplay_video(urlh.geturl(), locale, webpage) + + +class ViceShowIE(InfoExtractor): + IE_NAME = 'vice:show' + _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?show/(?P[^/?#&]+)' + + _TEST = { + 'url': 'https://munchies.vice.com/en/show/fuck-thats-delicious-2', + 'info_dict': { + 'id': 'fuck-thats-delicious-2', + 'title': "Fuck, That's Delicious", + 'description': 'Follow the culinary adventures of rapper Action Bronson during his ongoing world tour.', + }, + 'playlist_count': 17, } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - name = mobj.group('name') - webpage = self._download_webpage(url, name) - try: - ooyala_url = self._og_search_video_url(webpage) - except ExtractorError: - try: - embed_code = self._search_regex( - r'OO.Player.create\(\'ooyalaplayer\', \'(.+?)\'', webpage, - u'ooyala embed code') - ooyala_url = OoyalaIE._url_for_embed_code(embed_code) - except ExtractorError: - raise ExtractorError(u'The page doesn\'t contain a video', expected=True) - return self.url_result(ooyala_url, ie='Ooyala') + show_id = self._match_id(url) + webpage = self._download_webpage(url, show_id) + + entries = [ + self.url_result(video_url, ViceIE.ie_key()) + for video_url, _ in re.findall( + r']+class="article-title"[^>]+data-id="\d+"[^>]*>\s*]+href="(%s.*?)"' + % ViceIE._VALID_URL, webpage)] + + title = self._search_regex( + r'(.+?)', webpage, 'title', default=None) + if title: + title = re.sub(r'(.+)\s*\|\s*.+$', r'\1', title).strip() + description = self._html_search_meta( + 'description', webpage, 'description') + + return self.playlist_result(entries, show_id, title, description) + + +class ViceArticleIE(InfoExtractor): + IE_NAME = 'vice:article' + _VALID_URL = r'https://www\.vice\.com/[^/]+/article/(?P[^?#]+)' + + _TESTS = [{ + 'url': 'https://www.vice.com/en_us/article/on-set-with-the-woman-making-mormon-porn-in-utah', + 'info_dict': { + 'id': '58dc0a3dee202d2a0ccfcbd8', + 'ext': 'mp4', + 'title': 'Mormon War on Porn ', + 'description': 'md5:ad396a2481e7f8afb5ed486878421090', + 'uploader': 'VICE', + 'uploader_id': '57a204088cb727dec794c693', + 'timestamp': 1489160690, + 'upload_date': '20170310', + }, + 'params': { + # AES-encrypted m3u8 + 'skip_download': True, + }, + 'add_ie': ['UplynkPreplay'], + }, { + 'url': 'https://www.vice.com/en_us/article/how-to-hack-a-car', + 'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2', + 'info_dict': { + 'id': '3jstaBeXgAs', + 'ext': 'mp4', + 'title': 'How to Hack a Car: Phreaked Out (Episode 2)', + 'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30', + 'uploader_id': 'MotherboardTV', + 'uploader': 'Motherboard', + 'upload_date': '20140529', + }, + 'add_ie': ['Youtube'], + }, { + 'url': 'https://www.vice.com/en_us/article/cowboy-capitalists-part-1', + 'only_matching': True, + }, { + 'url': 'https://www.vice.com/ru/article/big-night-out-ibiza-clive-martin-229', + 'only_matching': True, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + prefetch_data = self._parse_json(self._search_regex( + r'window\.__PREFETCH_DATA\s*=\s*({.*});', + webpage, 'prefetch data'), display_id) + body = prefetch_data['body'] + + def _url_res(video_url, ie_key): + return { + '_type': 'url_transparent', + 'url': video_url, + 'display_id': display_id, + 'ie_key': ie_key, + } + + embed_code = self._search_regex( + r'embedCode=([^&\'"]+)', body, + 'ooyala embed code', default=None) + if embed_code: + return _url_res('ooyala:%s' % embed_code, 'Ooyala') + + youtube_url = YoutubeIE._extract_url(body) + if youtube_url: + return _url_res(youtube_url, YoutubeIE.ie_key()) + + video_url = self._html_search_regex( + r'data-video-url="([^"]+)"', + prefetch_data['embed_code'], 'video URL') + return _url_res(video_url, ViceIE.ie_key())