X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/540fd68c40df72763aee5d75598675c45cfa9aba..1918d37666ea9185c49ca84e64f58e4ff2a422c9:/youtube_dl/extractor/francetv.py diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index edf555b..095bb39 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -1,31 +1,29 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals import re -import json from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( clean_html, ExtractorError, int_or_none, - float_or_none, parse_duration, determine_ext, ) +from .dailymotion import DailymotionIE class FranceTVBaseInfoExtractor(InfoExtractor): - def _extract_video(self, video_id, catalogue): + def _extract_video(self, video_id, catalogue=None): info = self._download_json( - 'http://webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=%s&catalogue=%s' - % (video_id, catalogue), - video_id, 'Downloading video 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( @@ -58,66 +56,169 @@ class FranceTVBaseInfoExtractor(InfoExtractor): # See https://github.com/rg3/youtube-dl/issues/3963 # m3u8 urls work fine continue - video_url_parsed = compat_urllib_parse_urlparse(video_url) f4m_url = self._download_webpage( - 'http://hdfauth.francetv.fr/esi/urltokengen2.html?url=%s' % video_url_parsed.path, + 'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url, video_id, 'Downloading f4m manifest token', fatal=False) if f4m_url: - formats.extend(self._extract_f4m_formats(f4m_url, video_id, 1, format_id)) + formats.extend(self._extract_f4m_formats( + f4m_url + '&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(video_url, video_id, 'mp4', m3u8_id=format_id)) + formats.extend(self._extract_m3u8_formats( + video_url, 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', - 'preference': 1, }) else: - formats.append({ - 'url': video_url, - 'format_id': format_id, - 'preference': -1, - }) + 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': info['titre'], + 'title': title, 'description': clean_html(info['synopsis']), 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']), - 'duration': float_or_none(info.get('real_duration'), 1000) or parse_duration(info['duree']), + 'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']), 'timestamp': int_or_none(info['diffusion']['timestamp']), 'formats': formats, + 'subtitles': subtitles, } -class PluzzIE(FranceTVBaseInfoExtractor): - IE_NAME = 'pluzz.francetv.fr' - _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html' +class FranceTVIE(FranceTVBaseInfoExtractor): + _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P[^/]+)\.html' - # Can't use tests, videos expire in 7 days + _TESTS = [{ + 'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html', + 'info_dict': { + 'id': '157550144', + 'ext': 'mp4', + 'title': '13h15, le dimanche... - Les mystères de Jésus', + 'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42', + 'timestamp': 1494156300, + 'upload_date': '20170507', + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + }, + }, { + # 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, + }] def _real_extract(self, url): - title = re.match(self._VALID_URL, url).group(1) - webpage = self._download_webpage(url, title) + 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, 'Pluzz') + 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('@') + return self._extract_video(video_id, catalogue) -class FranceTvInfoIE(FranceTVBaseInfoExtractor): + +class FranceTVEmbedIE(FranceTVBaseInfoExtractor): + _VALID_URL = r'https?://embed\.francetv\.fr/*\?.*?\bue=(?P[^&]+)' + + _TEST = { + '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, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + video = self._download_json( + 'http://api-embed.webservices.francetelevisions.fr/key/%s' % video_id, + video_id) + + return self._extract_video(video['video_id'], video.get('catalog')) + + +class FranceTVInfoIE(FranceTVBaseInfoExtractor): IE_NAME = 'francetvinfo.fr' - _VALID_URL = r'https?://(?:www|mobile)\.francetvinfo\.fr/.*/(?P.+)\.html' + _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<title>[^/?#&.]+)' _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': 'flv', + 'ext': 'mp4', 'title': 'Soir 3', 'upload_date': '20130826', 'timestamp': 1377548400, + 'subtitles': { + 'fr': 'mincount:2', + }, + }, + 'params': { + # m3u8 downloads + 'skip_download': True, }, }, { 'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html', @@ -131,128 +232,100 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): 'skip_download': 'HLS (reqires ffmpeg)' }, 'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.', - }] - - 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, catalogue = self._search_regex( - r'id-video=([^@]+@[^"]+)', webpage, 'video id').split('@') - return self._extract_video(video_id, catalogue) - - -class FranceTVIE(FranceTVBaseInfoExtractor): - IE_NAME = 'francetv' - IE_DESC = 'France 2, 3, 4, 5 and Ô' - _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/ - (?: - emissions/.*?/(videos|emissions)/(?P<id>[^/?]+) - | (emissions?|jt)/(?P<key>[^/?]+) - )''' - - _TESTS = [ - # france2 - { - 'url': 'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104', - 'md5': 'c03fc87cb85429ffd55df32b9fc05523', - 'info_dict': { - 'id': '109169362', - 'ext': 'flv', - 'title': '13h15, le dimanche...', - 'description': 'md5:9a0932bb465f22d377a449be9d1a0ff7', - 'upload_date': '20140914', - 'timestamp': 1410693600, - }, + }, { + 'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html', + 'md5': 'f485bda6e185e7d15dbc69b72bae993e', + 'info_dict': { + 'id': 'NI_173343', + 'ext': 'mp4', + 'title': 'Les entreprises familiales : le secret de la réussite', + 'thumbnail': r're:^https?://.*\.jpe?g$', + 'timestamp': 1433273139, + 'upload_date': '20150602', }, - # france3 - { - 'url': 'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575', - 'md5': '679bb8f8921f8623bd658fa2f8364da0', - 'info_dict': { - 'id': '000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au', - 'ext': 'mp4', - 'title': 'Le scandale du prix des médicaments', - 'description': 'md5:1384089fbee2f04fc6c9de025ee2e9ce', - 'upload_date': '20131113', - 'timestamp': 1384380000, - }, + 'params': { + # m3u8 downloads + 'skip_download': True, }, - # france4 - { - 'url': 'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4', - 'md5': 'a182bf8d2c43d88d46ec48fbdd260c1c', - 'info_dict': { - 'id': 'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4', - 'ext': 'mp4', - 'title': 'Hero Corp Making of - Extrait 1', - 'description': 'md5:c87d54871b1790679aec1197e73d650a', - 'upload_date': '20131106', - 'timestamp': 1383766500, - }, + }, { + 'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html', + 'md5': 'f485bda6e185e7d15dbc69b72bae993e', + 'info_dict': { + 'id': 'NI_657393', + 'ext': 'mp4', + 'title': 'Olivier Monthus, réalisateur de "Bretagne, le choix de l’Armor"', + 'description': 'md5:a3264114c9d29aeca11ced113c37b16c', + 'thumbnail': r're:^https?://.*\.jpe?g$', + 'timestamp': 1458300695, + 'upload_date': '20160318', }, - # france5 - { - 'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968', - 'md5': '78f0f4064f9074438e660785bbf2c5d9', - 'info_dict': { - 'id': '108961659', - 'ext': 'flv', - 'title': 'C à dire ?!', - 'description': 'md5:1a4aeab476eb657bf57c4ff122129f81', - 'upload_date': '20140915', - 'timestamp': 1410795000, - }, + 'params': { + 'skip_download': True, }, - # franceo - { - 'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013', - 'md5': '52f0bfe202848b15915a2f39aaa8981b', - 'info_dict': { - 'id': '108634970', - 'ext': 'flv', - 'title': 'Infô Afrique', - 'description': 'md5:ebf346da789428841bee0fd2a935ea55', - 'upload_date': '20140915', - 'timestamp': 1410822000, - }, + }, { + # 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) - webpage = self._download_webpage(url, mobj.group('key') or mobj.group('id')) - video_id, catalogue = self._html_search_regex( - r'href="http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', - webpage, 'video ID').split('@') + page_title = mobj.group('title') + webpage = self._download_webpage(url, page_title) + + 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._extract_video(video_id, catalogue) -class GenerationQuoiIE(InfoExtractor): - IE_NAME = 'france2.fr:generation-quoi' - _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<id>[^/?#]+)' +class GenerationWhatIE(InfoExtractor): + IE_NAME = 'france2.fr:generation-what' + _VALID_URL = r'https?://generation-what\.francetv\.fr/[^/]+/video/(?P<id>[^/?#]+)' - _TEST = { - 'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous', + _TESTS = [{ + 'url': 'http://generation-what.francetv.fr/portrait/video/present-arms', 'info_dict': { - 'id': 'k7FJX8VBcvvLmX4wA5Q', + 'id': 'wtvKYUG45iw', 'ext': 'mp4', - 'title': 'Génération Quoi - Garde à Vous', - 'uploader': 'Génération Quoi', + 'title': 'Generation What - Garde à vous - FRA', + 'uploader': 'Generation What', + 'uploader_id': 'UCHH9p1eetWCgt4kXBYCb3_w', + 'upload_date': '20160411', }, - 'params': { - # It uses Dailymotion - 'skip_download': True, - }, - } + }, { + 'url': 'http://generation-what.francetv.fr/europe/video/present-arms', + 'only_matching': True, + }] def _real_extract(self, url): display_id = self._match_id(url) - info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % display_id) - info_json = self._download_webpage(info_url, display_id) - info = json.loads(info_json) - return self.url_result('http://www.dailymotion.com/video/%s' % info['id'], - ie='Dailymotion') + 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, 'Youtube', youtube_id) class CultureboxIE(FranceTVBaseInfoExtractor): @@ -283,6 +356,7 @@ class CultureboxIE(FranceTVBaseInfoExtractor): raise ExtractorError('Video %s is not available' % name, expected=True) video_id, catalogue = self._search_regex( - r'"http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', webpage, 'video id').split('@') + r'["\'>]https?://videos\.francetv\.fr/video/([^@]+@.+?)["\'<]', + webpage, 'video id').split('@') return self._extract_video(video_id, catalogue)