X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9c8b63077a48f758bf0c9a7351669557071bbd74..903490db3cf588364c9ff08098b1a9193438a7f3:/youtube_dl/extractor/francetv.py diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index 086cafc..c02cd03 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -1,129 +1,490 @@ -# encoding: utf-8 +# coding: utf-8 + +from __future__ import unicode_literals + import re -import xml.etree.ElementTree -import json from .common import InfoExtractor -from ..utils import ( +from ..compat import ( + compat_str, compat_urlparse, ) +from ..utils import ( + clean_html, + determine_ext, + ExtractorError, + int_or_none, + parse_duration, + try_get, +) +from .dailymotion import DailymotionIE class FranceTVBaseInfoExtractor(InfoExtractor): - def _extract_video(self, video_id): - xml_desc = self._download_webpage( - 'http://www.francetvinfo.fr/appftv/webservices/video/' - 'getInfosOeuvre.php?id-diffusion=' - + video_id, video_id, 'Downloading XML config') - info = xml.etree.ElementTree.fromstring(xml_desc.encode('utf-8')) + def _make_url_result(self, video_or_full_id, catalog=None): + full_id = 'francetv:%s' % video_or_full_id + if '@' not in video_or_full_id and catalog: + full_id += '@%s' % catalog + return self.url_result( + full_id, ie=FranceTVIE.ie_key(), + video_id=video_or_full_id.split('@')[0]) - manifest_url = info.find('videos/video/url').text - video_url = manifest_url.replace('manifest.f4m', 'index_2_av.m3u8') - video_url = video_url.replace('/z/', '/i/') - thumbnail_path = info.find('image').text - return {'id': video_id, - 'ext': 'mp4', - 'url': video_url, - 'title': info.find('titre').text, - 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path), - 'description': info.find('synopsis').text, - } +class FranceTVIE(InfoExtractor): + _VALID_URL = r'''(?x) + (?: + https?:// + sivideo\.webservices\.francetelevisions\.fr/tools/getInfosOeuvre/v2/\? + .*?\bidDiffusion=[^&]+| + (?: + https?://videos\.francetv\.fr/video/| + francetv: + ) + (?P[^@]+)(?:@(?P.+))? + ) + ''' + _TESTS = [{ + # without catalog + 'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=162311093&callback=_jsonp_loader_callback_request_0', + 'md5': 'c2248a8de38c4e65ea8fae7b5df2d84f', + 'info_dict': { + 'id': '162311093', + 'ext': 'mp4', + 'title': '13h15, le dimanche... - Les mystères de Jésus', + 'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42', + 'timestamp': 1502623500, + 'upload_date': '20170813', + }, + }, { + # with catalog + 'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=NI_1004933&catalogue=Zouzous&callback=_jsonp_loader_callback_request_4', + 'only_matching': True, + }, { + 'url': 'http://videos.francetv.fr/video/NI_657393@Regions', + 'only_matching': True, + }, { + 'url': 'francetv:162311093', + 'only_matching': True, + }, { + 'url': 'francetv:NI_1004933@Zouzous', + 'only_matching': True, + }, { + 'url': 'francetv:NI_983319@Info-web', + 'only_matching': True, + }, { + 'url': 'francetv:NI_983319', + 'only_matching': True, + }, { + 'url': 'francetv:NI_657393@Regions', + 'only_matching': True, + }, { + # france-3 live + 'url': 'francetv:SIM_France3', + 'only_matching': True, + }] + + def _extract_video(self, video_id, catalogue=None): + # Videos are identified by idDiffusion so catalogue part is optional. + # However when provided, some extra formats may be returned so we pass + # it if available. + info = self._download_json( + 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/', + video_id, 'Downloading video JSON', query={ + 'idDiffusion': video_id, + 'catalogue': catalogue or '', + }) + + if info.get('status') == 'NOK': + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, info['message']), + expected=True) + allowed_countries = info['videos'][0].get('geoblocage') + if allowed_countries: + georestricted = True + geo_info = self._download_json( + 'http://geo.francetv.fr/ws/edgescape.json', video_id, + 'Downloading geo restriction info') + country = geo_info['reponse']['geo_info']['country_code'] + if country not in allowed_countries: + raise ExtractorError( + 'The video is not available from your location', + expected=True) + else: + georestricted = False -class PluzzIE(FranceTVBaseInfoExtractor): - IE_NAME = u'pluzz.francetv.fr' - _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html' + def sign(manifest_url, manifest_id): + for host in ('hdfauthftv-a.akamaihd.net', 'hdfauth.francetv.fr'): + signed_url = self._download_webpage( + 'https://%s/esi/TA' % host, video_id, + 'Downloading signed %s manifest URL' % manifest_id, + fatal=False, query={ + 'url': manifest_url, + }) + if (signed_url and isinstance(signed_url, compat_str) and + re.search(r'^(?:https?:)?//', signed_url)): + return signed_url + return manifest_url - # Can't use tests, videos expire in 7 days + is_live = None + + formats = [] + for video in info['videos']: + if video['statut'] != 'ONLINE': + continue + video_url = video['url'] + if not video_url: + continue + if is_live is None: + is_live = (try_get( + video, lambda x: x['plages_ouverture'][0]['direct'], + bool) is True) or '/live.francetv.fr/' in video_url + format_id = video['format'] + ext = determine_ext(video_url) + if ext == 'f4m': + if georestricted: + # See https://github.com/rg3/youtube-dl/issues/3963 + # m3u8 urls work fine + continue + formats.extend(self._extract_f4m_formats( + sign(video_url, format_id) + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44', + video_id, f4m_id=format_id, fatal=False)) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + sign(video_url, format_id), video_id, 'mp4', + entry_protocol='m3u8_native', m3u8_id=format_id, + fatal=False)) + elif video_url.startswith('rtmp'): + formats.append({ + 'url': video_url, + 'format_id': 'rtmp-%s' % format_id, + 'ext': 'flv', + }) + else: + if self._is_valid_url(video_url, video_id, format_id): + formats.append({ + 'url': video_url, + 'format_id': format_id, + }) + self._sort_formats(formats) + + title = info['titre'] + subtitle = info.get('sous_titre') + if subtitle: + title += ' - %s' % subtitle + title = title.strip() + + subtitles = {} + subtitles_list = [{ + 'url': subformat['url'], + 'ext': subformat.get('format'), + } for subformat in info.get('subtitles', []) if subformat.get('url')] + if subtitles_list: + subtitles['fr'] = subtitles_list + + return { + 'id': video_id, + 'title': self._live_title(title) if is_live else title, + 'description': clean_html(info['synopsis']), + 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']), + 'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']), + 'timestamp': int_or_none(info['diffusion']['timestamp']), + 'is_live': is_live, + 'formats': formats, + 'subtitles': subtitles, + } def _real_extract(self, url): - title = re.match(self._VALID_URL, url).group(1) - webpage = self._download_webpage(url, title) + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + catalog = mobj.group('catalog') + + if not video_id: + qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + video_id = qs.get('idDiffusion', [None])[0] + catalog = qs.get('catalogue', [None])[0] + if not video_id: + raise ExtractorError('Invalid URL', expected=True) + + return self._extract_video(video_id, catalog) + + +class FranceTVSiteIE(FranceTVBaseInfoExtractor): + _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P[^/]+)\.html' + + _TESTS = [{ + 'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html', + 'info_dict': { + 'id': '162311093', + 'ext': 'mp4', + 'title': '13h15, le dimanche... - Les mystères de Jésus', + 'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42', + 'timestamp': 1502623500, + 'upload_date': '20170813', + }, + 'params': { + 'skip_download': True, + }, + 'add_ie': [FranceTVIE.ie_key()], + }, { + # france3 + 'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html', + 'only_matching': True, + }, { + # france4 + 'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html', + 'only_matching': True, + }, { + # france5 + 'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html', + 'only_matching': True, + }, { + # franceo + 'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html', + 'only_matching': True, + }, { + # france2 live + 'url': 'https://www.france.tv/france-2/direct.html', + 'only_matching': True, + }, { + 'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html', + 'only_matching': True, + }, { + 'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html', + 'only_matching': True, + }, { + 'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html', + 'only_matching': True, + }, { + 'url': 'https://www.france.tv/142749-rouge-sang.html', + 'only_matching': True, + }, { + # france-3 live + 'url': 'https://www.france.tv/france-3/direct.html', + 'only_matching': True, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + catalogue = None video_id = self._search_regex( - r'data-diffusion="(\d+)"', webpage, 'ID') - return self._extract_video(video_id) + r'data-main-video=(["\'])(?P(?:(?!\1).)+)\1', + webpage, 'video id', default=None, group='id') + if not video_id: + video_id, catalogue = self._html_search_regex( + r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"', + webpage, 'video ID').split('@') -class FranceTvInfoIE(FranceTVBaseInfoExtractor): - IE_NAME = u'francetvinfo.fr' - _VALID_URL = r'https?://www\.francetvinfo\.fr/replay.*/(?P.+).html' + return self._make_url_result(video_id, catalogue) - _TEST = { - u'url': u'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html', - u'file': u'84981923.mp4', - u'info_dict': { - u'title': u'Soir 3', + +class FranceTVEmbedIE(FranceTVBaseInfoExtractor): + _VALID_URL = r'https?://embed\.francetv\.fr/*\?.*?\bue=(?P<id>[^&]+)' + + _TESTS = [{ + 'url': 'http://embed.francetv.fr/?ue=7fd581a2ccf59d2fc5719c5c13cf6961', + 'info_dict': { + 'id': 'NI_983319', + 'ext': 'mp4', + 'title': 'Le Pen Reims', + 'upload_date': '20170505', + 'timestamp': 1493981780, + 'duration': 16, }, - u'params': { - u'skip_download': True, + 'params': { + 'skip_download': True, }, - } + 'add_ie': [FranceTVIE.ie_key()], + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - page_title = mobj.group('title') - webpage = self._download_webpage(url, page_title) - video_id = self._search_regex(r'id-video=(\d+?)"', webpage, u'video id') - return self._extract_video(video_id) - - -class France2IE(FranceTVBaseInfoExtractor): - IE_NAME = u'france2.fr' - _VALID_URL = r'''(?x)https?://www\.france2\.fr/ - (?: - emissions/.*?/videos/(?P<id>\d+) - | emission/(?P<key>[^/?]+) - )''' - - _TEST = { - u'url': u'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104', - u'file': u'75540104.mp4', - u'info_dict': { - u'title': u'13h15, le samedi...', - u'description': u'md5:2e5b58ba7a2d3692b35c792be081a03d', + video_id = self._match_id(url) + + video = self._download_json( + 'http://api-embed.webservices.francetelevisions.fr/key/%s' % video_id, + video_id) + + return self._make_url_result(video['video_id'], video.get('catalog')) + + +class FranceTVInfoIE(FranceTVBaseInfoExtractor): + IE_NAME = 'francetvinfo.fr' + _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&.]+)' + + _TESTS = [{ + 'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html', + 'info_dict': { + 'id': '84981923', + 'ext': 'mp4', + 'title': 'Soir 3', + 'upload_date': '20130826', + 'timestamp': 1377548400, + 'subtitles': { + 'fr': 'mincount:2', + }, }, - u'params': { - u'skip_download': True, + 'params': { + 'skip_download': True, }, - } + 'add_ie': [FranceTVIE.ie_key()], + }, { + 'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html', + 'only_matching': True, + }, { + 'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html', + 'only_matching': True, + }, { + 'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html', + 'only_matching': True, + }, { + # Dailymotion embed + 'url': 'http://www.francetvinfo.fr/politique/notre-dame-des-landes/video-sur-france-inter-cecile-duflot-denonce-le-regard-meprisant-de-patrick-cohen_1520091.html', + 'md5': 'ee7f1828f25a648addc90cb2687b1f12', + 'info_dict': { + 'id': 'x4iiko0', + 'ext': 'mp4', + 'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen', + 'description': 'Au lendemain de la victoire du "oui" au référendum sur l\'aéroport de Notre-Dame-des-Landes, l\'ancienne ministre écologiste est l\'invitée de Patrick Cohen. Plus d\'info : https://www.franceinter.fr/emissions/le-7-9/le-7-9-27-juin-2016', + 'timestamp': 1467011958, + 'upload_date': '20160627', + 'uploader': 'France Inter', + 'uploader_id': 'x2q2ez', + }, + 'add_ie': ['Dailymotion'], + }, { + 'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin', + 'only_matching': True, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj.group('key'): - webpage = self._download_webpage(url, mobj.group('key')) - video_id = self._html_search_regex( - r'''(?x)<div\s+class="video-player">\s* - <a\s+href="http://videos.francetv.fr/video/([0-9]+)"\s+ - class="francetv-video-player">''', - webpage, u'video ID') - else: - video_id = mobj.group('id') - return self._extract_video(video_id) + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + dailymotion_urls = DailymotionIE._extract_urls(webpage) + if dailymotion_urls: + return self.playlist_result([ + self.url_result(dailymotion_url, DailymotionIE.ie_key()) + for dailymotion_url in dailymotion_urls]) + + video_id, catalogue = self._search_regex( + (r'id-video=([^@]+@[^"]+)', + r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"'), + webpage, 'video id').split('@') + + return self._make_url_result(video_id, catalogue) + + +class GenerationWhatIE(InfoExtractor): + IE_NAME = 'france2.fr:generation-what' + _VALID_URL = r'https?://generation-what\.francetv\.fr/[^/]+/video/(?P<id>[^/?#&]+)' + + _TESTS = [{ + 'url': 'http://generation-what.francetv.fr/portrait/video/present-arms', + 'info_dict': { + 'id': 'wtvKYUG45iw', + 'ext': 'mp4', + 'title': 'Generation What - Garde à vous - FRA', + 'uploader': 'Generation What', + 'uploader_id': 'UCHH9p1eetWCgt4kXBYCb3_w', + 'upload_date': '20160411', + }, + 'params': { + 'skip_download': True, + }, + 'add_ie': ['Youtube'], + }, { + 'url': 'http://generation-what.francetv.fr/europe/video/present-arms', + 'only_matching': True, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + youtube_id = self._search_regex( + r"window\.videoURL\s*=\s*'([0-9A-Za-z_-]{11})';", + webpage, 'youtube id') + return self.url_result(youtube_id, ie='Youtube', video_id=youtube_id) -class GenerationQuoiIE(InfoExtractor): - IE_NAME = u'france2.fr:generation-quoi' - _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<name>.*)(\?|$)' - _TEST = { - u'url': u'http://generation-quoi.france2.fr/portrait/garde-a-vous', - u'file': u'k7FJX8VBcvvLmX4wA5Q.mp4', - u'info_dict': { - u'title': u'Génération Quoi - Garde à Vous', - u'uploader': u'Génération Quoi', +class CultureboxIE(FranceTVBaseInfoExtractor): + _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)' + + _TESTS = [{ + 'url': 'https://culturebox.francetvinfo.fr/opera-classique/musique-classique/c-est-baroque/concerts/cantates-bwv-4-106-et-131-de-bach-par-raphael-pichon-57-268689', + 'info_dict': { + 'id': 'EV_134885', + 'ext': 'mp4', + 'title': 'Cantates BWV 4, 106 et 131 de Bach par Raphaël Pichon 5/7', + 'description': 'md5:19c44af004b88219f4daa50fa9a351d4', + 'upload_date': '20180206', + 'timestamp': 1517945220, + 'duration': 5981, + }, + 'params': { + 'skip_download': True, + }, + 'add_ie': [FranceTVIE.ie_key()], + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + if ">Ce live n'est plus disponible en replay<" in webpage: + raise ExtractorError( + 'Video %s is not available' % display_id, expected=True) + + video_id, catalogue = self._search_regex( + r'["\'>]https?://videos\.francetv\.fr/video/([^@]+@.+?)["\'<]', + webpage, 'video id').split('@') + + return self._make_url_result(video_id, catalogue) + + +class FranceTVJeunesseIE(FranceTVBaseInfoExtractor): + _VALID_URL = r'(?P<url>https?://(?:www\.)?(?:zouzous|ludo)\.fr/heros/(?P<id>[^/?#&]+))' + + _TESTS = [{ + 'url': 'https://www.zouzous.fr/heros/simon', + 'info_dict': { + 'id': 'simon', }, - u'params': { - # It uses Dailymotion - u'skip_download': True, + 'playlist_count': 9, + }, { + 'url': 'https://www.ludo.fr/heros/ninjago', + 'info_dict': { + 'id': 'ninjago', }, - } + 'playlist_count': 10, + }, { + 'url': 'https://www.zouzous.fr/heros/simon?abc', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - name = mobj.group('name') - info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % name) - info_json = self._download_webpage(info_url, name) - info = json.loads(info_json) - return self.url_result('http://www.dailymotion.com/video/%s' % info['id'], - ie='Dailymotion') + playlist_id = mobj.group('id') + + playlist = self._download_json( + '%s/%s' % (mobj.group('url'), 'playlist'), playlist_id) + + if not playlist.get('count'): + raise ExtractorError( + '%s is not available' % playlist_id, expected=True) + + entries = [] + for item in playlist['items']: + identity = item.get('identity') + if identity and isinstance(identity, compat_str): + entries.append(self._make_url_result(identity)) + + return self.playlist_result(entries, playlist_id)