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'http://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)/)?(?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 embed_html = self._parse_json(
115 r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n',
117 video_id)['embed_html']
118 iframe_url = find_iframe_url(embed_html)
119 json_url = compat_parse_qs(
120 compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
121 return self._extract_from_json_url(json_url, video_id, lang)
123 def _extract_from_json_url(self, json_url, video_id, lang):
124 info = self._download_json(json_url, video_id)
125 player_info = info['videoJsonPlayer']
127 upload_date_str = player_info.get('shootingDate')
128 if not upload_date_str:
129 upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
131 title = player_info['VTI'].strip()
132 subtitle = player_info.get('VSU', '').strip()
134 title += ' - %s' % subtitle
137 'id': player_info['VID'],
139 'description': player_info.get('VDE'),
140 'upload_date': unified_strdate(upload_date_str),
141 'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
143 qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
153 for format_id, format_dict in player_info['VSR'].items():
154 f = dict(format_dict)
155 versionCode = f.get('versionCode')
156 langcode = LANGS.get(lang, lang)
157 lang_rexs = [r'VO?%s-' % re.escape(langcode), r'VO?.-ST%s$' % re.escape(langcode)]
160 matched_lang_rexs = [r for r in lang_rexs if re.match(r, versionCode)]
161 lang_pref = -10 if not matched_lang_rexs else 10 * len(matched_lang_rexs)
163 if versionCode is not None:
164 # The original version with subtitles has lower relevance
165 if re.match(r'VO-ST(F|A|E)', versionCode):
167 # The version with sourds/mal subtitles has also lower relevance
168 elif re.match(r'VO?(F|A|E)-STM\1', versionCode):
171 'format_id': format_id,
172 'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
173 'language_preference': lang_pref,
174 'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
175 'width': int_or_none(f.get('width')),
176 'height': int_or_none(f.get('height')),
177 'tbr': int_or_none(f.get('bitrate')),
178 'quality': qfunc(f.get('quality')),
179 'source_preference': source_pref,
182 if f.get('mediaType') == 'rtmp':
183 format['url'] = f['streamer']
184 format['play_path'] = 'mp4:' + f['url']
185 format['ext'] = 'flv'
187 format['url'] = f['url']
189 formats.append(format)
191 self._check_formats(formats, video_id)
192 self._sort_formats(formats)
194 info_dict['formats'] = formats
198 # It also uses the arte_vp_url url from the webpage to extract the information
199 class ArteTVCreativeIE(ArteTVPlus7IE):
200 IE_NAME = 'arte.tv:creative'
201 _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:magazine?/)?(?P<id>[^?#]+)'
204 'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
208 'title': 'Folge 2 - Corporate Design',
209 'upload_date': '20131004',
212 'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
216 'title': 'Monty Python live (mostly)',
217 'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
218 'upload_date': '20140805',
223 class ArteTVFutureIE(ArteTVPlus7IE):
224 IE_NAME = 'arte.tv:future'
225 _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
228 'url': 'http://future.arte.tv/fr/info-sciences/les-ecrevisses-aussi-sont-anxieuses',
230 'id': '050940-028-A',
232 'title': 'Les écrevisses aussi peuvent être anxieuses',
235 'url': 'http://future.arte.tv/fr/la-science-est-elle-responsable',
236 'only_matching': True,
240 class ArteTVDDCIE(ArteTVPlus7IE):
241 IE_NAME = 'arte.tv:ddc'
242 _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
244 def _real_extract(self, url):
245 video_id, lang = self._extract_url_info(url)
248 elif lang == 'emission':
250 webpage = self._download_webpage(url, video_id)
251 scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
252 script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
253 javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
254 json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
255 return self._extract_from_json_url(json_url, video_id, lang)
258 class ArteTVConcertIE(ArteTVPlus7IE):
259 IE_NAME = 'arte.tv:concert'
260 _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
263 'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
264 'md5': '9ea035b7bd69696b67aa2ccaaa218161',
268 'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
269 'upload_date': '20140128',
270 'description': 'md5:486eb08f991552ade77439fe6d82c305',
275 class ArteTVCinemaIE(ArteTVPlus7IE):
276 IE_NAME = 'arte.tv:cinema'
277 _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
280 'url': 'http://cinema.arte.tv/de/node/38291',
281 'md5': '6b275511a5107c60bacbeeda368c3aa1',
283 'id': '055876-000_PWA12025-D',
285 'title': 'Tod auf dem Nil',
286 'upload_date': '20160122',
287 'description': 'md5:7f749bbb77d800ef2be11d54529b96bc',
292 class ArteTVMagazineIE(ArteTVPlus7IE):
293 IE_NAME = 'arte.tv:magazine'
294 _VALID_URL = r'https?://(?:www\.)?arte\.tv/magazine/(?P<name>.*?)/(?P<lang>fr|de)/(?P<id>.*?)'
297 'url': 'http://www.arte.tv/magazine/trepalium/fr/entretien-avec-le-realisateur-vincent-lannoo-trepalium',
298 'md5': '66a093339c1278bb3719157ef07107b2',
300 'id': '065965-000-A',
302 'title': 'Trepalium - Extrait Ep.01',
305 'url': 'http://www.arte.tv/magazine/metropolis/de/frank-woeste-german-paris-metropolis',
306 'md5': '53185a2b4876383552c037c21aeca8f7',
308 'id': '066149-000-A',
310 'title': 'Bonus: Frank Woeste - ein deutscher Jazzmusiker in Paris',
315 class ArteTVEmbedIE(ArteTVPlus7IE):
316 IE_NAME = 'arte.tv:embed'
317 _VALID_URL = r'''(?x)
319 /playerv2/embed\.php\?json_url=
321 http://arte\.tv/papi/tvguide/videos/stream/player/
322 (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*
326 def _real_extract(self, url):
327 mobj = re.match(self._VALID_URL, url)
328 video_id = mobj.group('id')
329 lang = mobj.group('lang')
330 json_url = mobj.group('json_url')
331 return self._extract_from_json_url(json_url, video_id, lang)