X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/6d835d4d6903cf5e18ad01844736929e06d16004..14111bdbc4a22585fc51fd4efa3572ea003b5321:/youtube_dl/extractor/francetv.py diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index 35d7d15..8e60cf6 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -6,14 +6,15 @@ import re import json from .common import InfoExtractor +from ..compat import compat_urlparse from ..utils import ( - compat_urlparse, - ExtractorError, clean_html, - parse_duration, - compat_urllib_parse_urlparse, + ExtractorError, int_or_none, + parse_duration, + determine_ext, ) +from .dailymotion import DailymotionCloudIE class FranceTVBaseInfoExtractor(InfoExtractor): @@ -26,6 +27,19 @@ class FranceTVBaseInfoExtractor(InfoExtractor): 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 formats = [] for video in info['videos']: @@ -35,18 +49,20 @@ class FranceTVBaseInfoExtractor(InfoExtractor): if not video_url: continue format_id = video['format'] - if video_url.endswith('.f4m'): - video_url_parsed = compat_urllib_parse_urlparse(video_url) + ext = determine_ext(video_url) + if ext == 'f4m': + if georestricted: + # See https://github.com/rg3/youtube-dl/issues/3963 + # m3u8 urls work fine + continue 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: - f4m_formats = self._extract_f4m_formats(f4m_url, video_id) - for f4m_format in f4m_formats: - f4m_format['preference'] = 1 - formats.extend(f4m_formats) - elif video_url.endswith('.m3u8'): - formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4')) + formats.extend(self._extract_f4m_formats( + f4m_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, 1, format_id)) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4', m3u8_id=format_id)) elif video_url.startswith('rtmp'): formats.append({ 'url': video_url, @@ -62,28 +78,48 @@ class FranceTVBaseInfoExtractor(InfoExtractor): }) self._sort_formats(formats) + title = info['titre'] + subtitle = info.get('sous_titre') + if subtitle: + title += ' - %s' % subtitle + + 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': 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' + _VALID_URL = r'https?://(?:m\.)?pluzz\.francetv\.fr/videos/(?P.+?)\.html' # Can't use tests, videos expire in 7 days def _real_extract(self, url): - title = re.match(self._VALID_URL, url).group(1) - webpage = self._download_webpage(url, title) - video_id = self._search_regex( - r'data-diffusion="(\d+)"', webpage, 'ID') + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + video_id = self._html_search_meta( + 'id_video', webpage, 'video id', default=None) + if not video_id: + video_id = self._search_regex( + r'data-diffusion=["\'](\d+)', webpage, 'video id') + return self._extract_video(video_id, 'Pluzz') @@ -99,6 +135,9 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): 'title': 'Soir 3', 'upload_date': '20130826', 'timestamp': 1377548400, + 'subtitles': { + 'fr': 'mincount:2', + }, }, }, { 'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html', @@ -112,12 +151,26 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): 'skip_download': 'HLS (reqires ffmpeg)' }, 'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.', + }, { + 'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html', + 'md5': 'f485bda6e185e7d15dbc69b72bae993e', + 'info_dict': { + 'id': '556e03339473995ee145930c', + 'ext': 'mp4', + 'title': 'Les entreprises familiales : le secret de la réussite', + 'thumbnail': 're:^https?://.*\.jpe?g$', + } }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) page_title = mobj.group('title') webpage = self._download_webpage(url, page_title) + + dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage) + if dmcloud_url: + return self.url_result(dmcloud_url, 'DailymotionCloud') + video_id, catalogue = self._search_regex( r'id-video=([^@]+@[^"]+)', webpage, 'video id').split('@') return self._extract_video(video_id, catalogue) @@ -126,11 +179,21 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): 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[^/?]+) - | (emissions?|jt)/(?P[^/?]+) - )''' + _VALID_URL = r'''(?x) + https?:// + (?: + (?:www\.)?france[2345o]\.fr/ + (?: + emissions/[^/]+/(?:videos|diffusions)| + emission/[^/]+| + videos| + jt + ) + /| + embed\.francetv\.fr/\?ue= + ) + (?P[^/?]+) + ''' _TESTS = [ # france2 @@ -174,49 +237,72 @@ class FranceTVIE(FranceTVBaseInfoExtractor): }, # france5 { - 'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968', - 'md5': '78f0f4064f9074438e660785bbf2c5d9', + 'url': 'http://www.france5.fr/emissions/c-a-dire/videos/quels_sont_les_enjeux_de_cette_rentree_politique__31-08-2015_908948?onglet=tous&page=1', + 'md5': 'f6c577df3806e26471b3d21631241fd0', 'info_dict': { - 'id': '108961659', + 'id': '123327454', 'ext': 'flv', - 'title': 'C à dire ?!', - 'description': 'md5:1a4aeab476eb657bf57c4ff122129f81', - 'upload_date': '20140915', - 'timestamp': 1410795000, + 'title': 'C à dire ?! - Quels sont les enjeux de cette rentrée politique ?', + 'description': 'md5:4a0d5cb5dce89d353522a84462bae5a4', + 'upload_date': '20150831', + 'timestamp': 1441035120, }, }, # franceo { - 'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013', - 'md5': '52f0bfe202848b15915a2f39aaa8981b', + 'url': 'http://www.franceo.fr/jt/info-soir/18-07-2015', + 'md5': '47d5816d3b24351cdce512ad7ab31da8', 'info_dict': { - 'id': '108634970', + 'id': '125377621', 'ext': 'flv', - 'title': 'Infô Afrique', - 'description': 'md5:ebf346da789428841bee0fd2a935ea55', - 'upload_date': '20140915', - 'timestamp': 1410822000, + 'title': 'Infô soir', + 'description': 'md5:01b8c6915a3d93d8bbbd692651714309', + 'upload_date': '20150718', + 'timestamp': 1437241200, + 'duration': 414, }, }, + { + # francetv embed + 'url': 'http://embed.francetv.fr/?ue=8d7d3da1e3047c42ade5a5d7dfd3fc87', + 'info_dict': { + 'id': 'EV_30231', + 'ext': 'flv', + 'title': 'Alcaline, le concert avec Calogero', + 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff', + 'upload_date': '20150226', + 'timestamp': 1424989860, + 'duration': 5400, + }, + }, + { + 'url': 'http://www.france4.fr/emission/highlander/diffusion-du-17-07-2015-04h05', + 'only_matching': True, + }, + { + 'url': 'http://www.franceo.fr/videos/125377617', + '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 = self._match_id(url) + webpage = self._download_webpage(url, video_id) video_id, catalogue = self._html_search_regex( - r'href="http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', + r'href="http://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.*)(\?|$)' + _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P[^/?#]+)' _TEST = { 'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous', - 'file': 'k7FJX8VBcvvLmX4wA5Q.mp4', 'info_dict': { + 'id': 'k7FJX8VBcvvLmX4wA5Q', + 'ext': 'mp4', 'title': 'Génération Quoi - Garde à Vous', 'uploader': 'Génération Quoi', }, @@ -224,17 +310,15 @@ class GenerationQuoiIE(InfoExtractor): # It uses Dailymotion 'skip_download': True, }, - 'skip': 'Only available from France', } 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) + 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') + ie='Dailymotion') class CultureboxIE(FranceTVBaseInfoExtractor): @@ -242,22 +326,28 @@ class CultureboxIE(FranceTVBaseInfoExtractor): _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?P.*?)(\?|$)' _TEST = { - 'url': 'http://culturebox.francetvinfo.fr/festivals/dans-les-jardins-de-william-christie/dans-les-jardins-de-william-christie-le-camus-162553', - 'md5': '5ad6dec1ffb2a3fbcb20cc4b744be8d6', + 'url': 'http://culturebox.francetvinfo.fr/live/musique/musique-classique/le-livre-vermeil-de-montserrat-a-la-cathedrale-delne-214511', + 'md5': '9b88dc156781c4dbebd4c3e066e0b1d6', 'info_dict': { - 'id': 'EV_22853', + 'id': 'EV_50111', 'ext': 'flv', - 'title': 'Dans les jardins de William Christie - Le Camus', - 'description': 'md5:4710c82315c40f0c865ca8b9a68b5299', - 'upload_date': '20140829', - 'timestamp': 1409317200, + 'title': "Le Livre Vermeil de Montserrat à la Cathédrale d'Elne", + 'description': 'md5:f8a4ad202e8fe533e2c493cc12e739d9', + 'upload_date': '20150320', + 'timestamp': 1426892400, + 'duration': 2760.9, }, } def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) name = mobj.group('name') + webpage = self._download_webpage(url, name) + + if ">Ce live n'est plus disponible en replay<" in webpage: + 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('@')