+ 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])
+
+
+class FranceTVIE(InfoExtractor):
+ _VALID_URL = r'''(?x)
+ (?:
+ https?://
+ sivideo\.webservices\.francetelevisions\.fr/tools/getInfosOeuvre/v2/\?
+ .*?\bidDiffusion=[^&]+|
+ (?:
+ https?://videos\.francetv\.fr/video/|
+ francetv:
+ )
+ (?P<id>[^@]+)(?:@(?P<catalog>.+))?
+ )
+ '''
+
+ _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
+
+ def sign(manifest_url, manifest_id):
+ for host in ('hdfauthftv-a.akamaihd.net', 'hdfauth.francetv.fr'):
+ signed_url = url_or_none(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:
+ return signed_url
+ return manifest_url
+
+ 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):
+ mobj = re.match(self._VALID_URL, url)
+ video_id = mobj.group('id')
+ catalog = mobj.group('catalog')