2 from __future__ import unicode_literals
6 from .common import InfoExtractor
9 compat_urllib_parse_urlparse,
14 get_element_by_attribute,
20 # There are different sources of video in arte.tv, the extraction process
21 # is different for each one. The videos usually expire in 7 days, so we can't
25 class ArteTvIE(InfoExtractor):
26 _VALID_URL = r'https?://videos\.arte\.tv/(?P<lang>fr|de|en|es)/.*-(?P<id>.*?)\.html'
29 def _real_extract(self, url):
30 mobj = re.match(self._VALID_URL, url)
31 lang = mobj.group('lang')
32 video_id = mobj.group('id')
34 ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
35 ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
36 ref_xml_doc = self._download_xml(
37 ref_xml_url, video_id, note='Downloading metadata')
38 config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
39 config_xml_url = config_node.attrib['ref']
40 config = self._download_xml(
41 config_xml_url, video_id, note='Downloading configuration')
44 'format_id': q.attrib['quality'],
45 # The playpath starts at 'mp4:', if we don't manually
46 # split the url, rtmpdump will incorrectly parse them
47 'url': q.text.split('mp4:', 1)[0],
48 'play_path': 'mp4:' + q.text.split('mp4:', 1)[1],
50 'quality': 2 if q.attrib['quality'] == 'hd' else 1,
51 } for q in config.findall('./urls/url')]
52 self._sort_formats(formats)
54 title = config.find('.//name').text
55 thumbnail = config.find('.//firstThumbnailUrl').text
59 'thumbnail': thumbnail,
64 class ArteTVPlus7IE(InfoExtractor):
65 IE_NAME = 'arte.tv:+7'
66 _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&+])'
69 def _extract_url_info(cls, url):
70 mobj = re.match(cls._VALID_URL, url)
71 lang = mobj.group('lang')
72 query = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
74 video_id = query['vid'][0]
76 # This is not a real id, it can be for example AJT for the news
77 # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
78 video_id = mobj.group('id')
81 def _real_extract(self, url):
82 video_id, lang = self._extract_url_info(url)
83 webpage = self._download_webpage(url, video_id)
84 return self._extract_from_webpage(webpage, video_id, lang)
86 def _extract_from_webpage(self, webpage, video_id, lang):
87 patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']')
89 # some pages contain multiple videos (like
90 # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D),
91 # so we first try to look for json URLs that contain the video id from
92 # the 'vid' parameter.
93 patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates]
94 json_url = self._html_search_regex(
95 patterns, webpage, 'json vp url', default=None)
97 def find_iframe_url(webpage, default=NO_DEFAULT):
98 return self._html_search_regex(
99 r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
100 webpage, 'iframe url', group='url', default=default)
102 iframe_url = find_iframe_url(webpage, None)
104 embed_url = self._html_search_regex(
105 r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url', default=None)
107 player = self._download_json(
108 embed_url, video_id, 'Downloading player page')
109 iframe_url = find_iframe_url(player['html'])
110 # en and es URLs produce react-based pages with different layout (e.g.
111 # http://www.arte.tv/guide/en/053330-002-A/carnival-italy?zone=world)
113 program = self._search_regex(
114 r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n',
115 webpage, 'program', default=None)
117 embed_html = self._parse_json(program, video_id)
119 iframe_url = find_iframe_url(embed_html['embed_html'])
121 json_url = compat_parse_qs(
122 compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
124 title = self._search_regex(
125 r'<h3[^>]+title=(["\'])(?P<title>.+?)\1',
126 webpage, 'title', default=None, group='title')
127 return self._extract_from_json_url(json_url, video_id, lang, title=title)
128 # Different kind of embed URL (e.g.
129 # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium)
130 embed_url = self._search_regex(
131 r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1',
132 webpage, 'embed url', group='url')
133 return self.url_result(embed_url)
135 def _extract_from_json_url(self, json_url, video_id, lang, title=None):
136 info = self._download_json(json_url, video_id)
137 player_info = info['videoJsonPlayer']
139 upload_date_str = player_info.get('shootingDate')
140 if not upload_date_str:
141 upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
143 title = (player_info.get('VTI') or title or player_info['VID']).strip()
144 subtitle = player_info.get('VSU', '').strip()
146 title += ' - %s' % subtitle
149 'id': player_info['VID'],
151 'description': player_info.get('VDE'),
152 'upload_date': unified_strdate(upload_date_str),
153 'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
155 qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
165 for format_id, format_dict in player_info['VSR'].items():
166 f = dict(format_dict)
167 versionCode = f.get('versionCode')
168 langcode = LANGS.get(lang, lang)
169 lang_rexs = [r'VO?%s-' % re.escape(langcode), r'VO?.-ST%s$' % re.escape(langcode)]
172 matched_lang_rexs = [r for r in lang_rexs if re.match(r, versionCode)]
173 lang_pref = -10 if not matched_lang_rexs else 10 * len(matched_lang_rexs)
175 if versionCode is not None:
176 # The original version with subtitles has lower relevance
177 if re.match(r'VO-ST(F|A|E)', versionCode):
179 # The version with sourds/mal subtitles has also lower relevance
180 elif re.match(r'VO?(F|A|E)-STM\1', versionCode):
183 'format_id': format_id,
184 'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
185 'language_preference': lang_pref,
186 'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
187 'width': int_or_none(f.get('width')),
188 'height': int_or_none(f.get('height')),
189 'tbr': int_or_none(f.get('bitrate')),
190 'quality': qfunc(f.get('quality')),
191 'source_preference': source_pref,
194 if f.get('mediaType') == 'rtmp':
195 format['url'] = f['streamer']
196 format['play_path'] = 'mp4:' + f['url']
197 format['ext'] = 'flv'
199 format['url'] = f['url']
201 formats.append(format)
203 self._check_formats(formats, video_id)
204 self._sort_formats(formats)
206 info_dict['formats'] = formats
210 # It also uses the arte_vp_url url from the webpage to extract the information
211 class ArteTVCreativeIE(ArteTVPlus7IE):
212 IE_NAME = 'arte.tv:creative'
213 _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:magazine?/)?(?P<id>[^/?#&]+)'
216 'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
220 'title': 'Folge 2 - Corporate Design',
221 'upload_date': '20131004',
224 'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
228 'title': 'Monty Python live (mostly)',
229 'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
230 'upload_date': '20140805',
235 class ArteTVFutureIE(ArteTVPlus7IE):
236 IE_NAME = 'arte.tv:future'
237 _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
240 'url': 'http://future.arte.tv/fr/info-sciences/les-ecrevisses-aussi-sont-anxieuses',
242 'id': '050940-028-A',
244 'title': 'Les écrevisses aussi peuvent être anxieuses',
245 'upload_date': '20140902',
248 'url': 'http://future.arte.tv/fr/la-science-est-elle-responsable',
249 'only_matching': True,
253 class ArteTVDDCIE(ArteTVPlus7IE):
254 IE_NAME = 'arte.tv:ddc'
255 _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>[^/?#&]+)'
257 def _real_extract(self, url):
258 video_id, lang = self._extract_url_info(url)
261 elif lang == 'emission':
263 webpage = self._download_webpage(url, video_id)
264 scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
265 script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
266 javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
267 json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
268 return self._extract_from_json_url(json_url, video_id, lang)
271 class ArteTVConcertIE(ArteTVPlus7IE):
272 IE_NAME = 'arte.tv:concert'
273 _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
276 'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
277 'md5': '9ea035b7bd69696b67aa2ccaaa218161',
281 'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
282 'upload_date': '20140128',
283 'description': 'md5:486eb08f991552ade77439fe6d82c305',
288 class ArteTVCinemaIE(ArteTVPlus7IE):
289 IE_NAME = 'arte.tv:cinema'
290 _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
293 'url': 'http://cinema.arte.tv/de/node/38291',
294 'md5': '6b275511a5107c60bacbeeda368c3aa1',
296 'id': '055876-000_PWA12025-D',
298 'title': 'Tod auf dem Nil',
299 'upload_date': '20160122',
300 'description': 'md5:7f749bbb77d800ef2be11d54529b96bc',
305 class ArteTVMagazineIE(ArteTVPlus7IE):
306 IE_NAME = 'arte.tv:magazine'
307 _VALID_URL = r'https?://(?:www\.)?arte\.tv/magazine/[^/]+/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
310 # Embedded via <iframe src="http://www.arte.tv/arte_vp/index.php?json_url=..."
311 'url': 'http://www.arte.tv/magazine/trepalium/fr/entretien-avec-le-realisateur-vincent-lannoo-trepalium',
312 'md5': '2a9369bcccf847d1c741e51416299f25',
314 'id': '065965-000-A',
316 'title': 'Trepalium - Extrait Ep.01',
317 'upload_date': '20160121',
320 # Embedded via <iframe src="http://www.arte.tv/guide/fr/embed/054813-004-A/medium"
321 'url': 'http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium',
322 'md5': 'fedc64fc7a946110fe311634e79782ca',
324 'id': '054813-004_PLUS7-F',
326 'title': 'Trepalium (4/6)',
327 'description': 'md5:10057003c34d54e95350be4f9b05cb40',
328 'upload_date': '20160218',
331 'url': 'http://www.arte.tv/magazine/metropolis/de/frank-woeste-german-paris-metropolis',
332 'only_matching': True,
336 class ArteTVEmbedIE(ArteTVPlus7IE):
337 IE_NAME = 'arte.tv:embed'
338 _VALID_URL = r'''(?x)
340 /playerv2/embed\.php\?json_url=
342 http://arte\.tv/papi/tvguide/videos/stream/player/
343 (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*
347 def _real_extract(self, url):
348 mobj = re.match(self._VALID_URL, url)
349 video_id = mobj.group('id')
350 lang = mobj.group('lang')
351 json_url = mobj.group('json_url')
352 return self._extract_from_json_url(json_url, video_id, lang)