X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/bc5059672e222ec4df4fe3983c399c0d0fca84d5..dd50658396011c8e98b7201f035408b7959d56e1:/youtube_dl/extractor/toutv.py diff --git a/youtube_dl/extractor/toutv.py b/youtube_dl/extractor/toutv.py index 1e9598e..2e7876c 100644 --- a/youtube_dl/extractor/toutv.py +++ b/youtube_dl/extractor/toutv.py @@ -1,71 +1,120 @@ # coding: utf-8 +from __future__ import unicode_literals + import re from .common import InfoExtractor from ..utils import ( - ExtractorError, - unified_strdate, + int_or_none, + js_to_json, + urlencode_postdata, + extract_attributes, + smuggle_url, ) class TouTvIE(InfoExtractor): - IE_NAME = u'tou.tv' - _VALID_URL = r'https?://www\.tou\.tv/(?P[a-zA-Z0-9_-]+(?:/(?PS[0-9]+E[0-9]+)))' + _NETRC_MACHINE = 'toutv' + IE_NAME = 'tou.tv' + _VALID_URL = r'https?://ici\.tou\.tv/(?P[a-zA-Z0-9_-]+(?:/S[0-9]+[EC][0-9]+)?)' + _access_token = None + _claims = None - _TEST = { - u'url': u'http://www.tou.tv/30-vies/S04E41', - u'file': u'30-vies_S04E41.mp4', - u'info_dict': { - u'title': u'30 vies Saison 4 / Épisode 41', - u'description': u'md5:da363002db82ccbe4dafeb9cab039b09', - u'age_limit': 8, - u'uploader': u'Groupe des Nouveaux Médias', - u'duration': 1296, - u'upload_date': u'20131118', - u'thumbnail': u'http://static.tou.tv/medias/images/2013-11-18_19_00_00_30VIES_0341_01_L.jpeg', + _TESTS = [{ + 'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17', + 'info_dict': { + 'id': '122017', + 'ext': 'mp4', + 'title': 'Saison 2015 Épisode 17', + 'description': 'La photo de famille 2', + 'upload_date': '20100717', }, - u'params': { - u'skip_download': True, # Requires rtmpdump + 'params': { + # m3u8 download + 'skip_download': True, }, - u'skip': 'Only available in Canada' - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - webpage = self._download_webpage(url, video_id) + 'skip': '404 Not Found', + }, { + 'url': 'http://ici.tou.tv/hackers', + 'only_matching': True, + }, { + 'url': 'https://ici.tou.tv/l-age-adulte/S01C501', + 'only_matching': True, + }] - mediaId = self._search_regex( - r'"idMedia":\s*"([^"]+)"', webpage, u'media ID') + def _real_initialize(self): + email, password = self._get_login_info() + if email is None: + return + state = 'http://ici.tou.tv/' + webpage = self._download_webpage(state, None, 'Downloading homepage') + toutvlogin = self._parse_json(self._search_regex( + r'(?s)toutvlogin\s*=\s*({.+?});', webpage, 'toutvlogin'), None, js_to_json) + authorize_url = toutvlogin['host'] + '/auth/oauth/v2/authorize' + login_webpage = self._download_webpage( + authorize_url, None, 'Downloading login page', query={ + 'client_id': toutvlogin['clientId'], + 'redirect_uri': 'https://ici.tou.tv/login/loginCallback', + 'response_type': 'token', + 'scope': 'media-drmt openid profile email id.write media-validation.read.privileged', + 'state': state, + }) - streams_url = u'http://release.theplatform.com/content.select?pid=' + mediaId - streams_doc = self._download_xml( - streams_url, video_id, note=u'Downloading stream list') + def extract_form_url_and_data(wp, default_form_url, form_spec_re=''): + form, form_elem = re.search( + r'(?s)((]+?%s[^>]*?>).+?)' % form_spec_re, wp).groups() + form_data = self._hidden_inputs(form) + form_url = extract_attributes(form_elem).get('action') or default_form_url + return form_url, form_data - video_url = next(n.text - for n in streams_doc.findall('.//choice/url') - if u'//ad.doubleclick' not in n.text) - if video_url.endswith('/Unavailable.flv'): - raise ExtractorError( - u'Access to this video is blocked from outside of Canada', - expected=True) + post_url, form_data = extract_form_url_and_data( + login_webpage, + 'https://services.radio-canada.ca/auth/oauth/v2/authorize/login', + r'(?:id|name)="Form-login"') + form_data.update({ + 'login-email': email, + 'login-password': password, + }) + consent_webpage = self._download_webpage( + post_url, None, 'Logging in', data=urlencode_postdata(form_data)) + post_url, form_data = extract_form_url_and_data( + consent_webpage, + 'https://services.radio-canada.ca/auth/oauth/v2/authorize/consent') + _, urlh = self._download_webpage_handle( + post_url, None, 'Following Redirection', + data=urlencode_postdata(form_data)) + self._access_token = self._search_regex( + r'access_token=([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})', + urlh.geturl(), 'access token') + self._claims = self._download_json( + 'https://services.radio-canada.ca/media/validation/v2/getClaims', + None, 'Extracting Claims', query={ + 'token': self._access_token, + 'access_token': self._access_token, + })['claims'] - duration_str = self._html_search_meta( - 'video:duration', webpage, u'duration') - duration = int(duration_str) if duration_str else None - upload_date_str = self._html_search_meta( - 'video:release_date', webpage, u'upload date') - upload_date = unified_strdate(upload_date_str) if upload_date_str else None + def _real_extract(self, url): + path = self._match_id(url) + metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path) + # IsDrm does not necessarily mean the video is DRM protected (see + # https://github.com/rg3/youtube-dl/issues/13994). + if metadata.get('IsDrm'): + self.report_warning('This video is probably DRM protected.', path) + video_id = metadata['IdMedia'] + details = metadata['Details'] + title = details['OriginalTitle'] + video_url = 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id) + if self._access_token and self._claims: + video_url = smuggle_url(video_url, { + 'access_token': self._access_token, + 'claims': self._claims, + }) return { - 'id': video_id, - 'title': self._og_search_title(webpage), + '_type': 'url_transparent', 'url': video_url, - 'description': self._og_search_description(webpage), - 'uploader': self._dc_search_uploader(webpage), - 'thumbnail': self._og_search_thumbnail(webpage), - 'age_limit': self._media_rating_search(webpage), - 'duration': duration, - 'upload_date': upload_date, - 'ext': 'mp4', + 'id': video_id, + 'title': title, + 'thumbnail': details.get('ImageUrl'), + 'duration': int_or_none(details.get('LengthInSeconds')), }