2 from __future__ import unicode_literals
6 from .common import InfoExtractor
12 get_element_by_attribute,
17 # There are different sources of video in arte.tv, the extraction process
18 # is different for each one. The videos usually expire in 7 days, so we can't
22 class ArteTvIE(InfoExtractor):
23 _VALID_URL = r'http://videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html'
26 def _real_extract(self, url):
27 mobj = re.match(self._VALID_URL, url)
28 lang = mobj.group('lang')
29 video_id = mobj.group('id')
31 ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
32 ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
33 ref_xml_doc = self._download_xml(
34 ref_xml_url, video_id, note='Downloading metadata')
35 config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
36 config_xml_url = config_node.attrib['ref']
37 config = self._download_xml(
38 config_xml_url, video_id, note='Downloading configuration')
41 'forma_id': q.attrib['quality'],
42 # The playpath starts at 'mp4:', if we don't manually
43 # split the url, rtmpdump will incorrectly parse them
44 'url': q.text.split('mp4:', 1)[0],
45 'play_path': 'mp4:' + q.text.split('mp4:', 1)[1],
47 'quality': 2 if q.attrib['quality'] == 'hd' else 1,
48 } for q in config.findall('./urls/url')]
49 self._sort_formats(formats)
51 title = config.find('.//name').text
52 thumbnail = config.find('.//firstThumbnailUrl').text
56 'thumbnail': thumbnail,
61 class ArteTVPlus7IE(InfoExtractor):
62 IE_NAME = 'arte.tv:+7'
63 _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
66 def _extract_url_info(cls, url):
67 mobj = re.match(cls._VALID_URL, url)
68 lang = mobj.group('lang')
69 # This is not a real id, it can be for example AJT for the news
70 # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
71 video_id = mobj.group('id')
74 def _real_extract(self, url):
75 video_id, lang = self._extract_url_info(url)
76 webpage = self._download_webpage(url, video_id)
77 return self._extract_from_webpage(webpage, video_id, lang)
79 def _extract_from_webpage(self, webpage, video_id, lang):
80 json_url = self._html_search_regex(
81 [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'],
82 webpage, 'json vp url')
83 return self._extract_from_json_url(json_url, video_id, lang)
85 def _extract_from_json_url(self, json_url, video_id, lang):
86 info = self._download_json(json_url, video_id)
87 player_info = info['videoJsonPlayer']
89 upload_date_str = player_info.get('shootingDate')
90 if not upload_date_str:
91 upload_date_str = player_info.get('VDA', '').split(' ')[0]
93 title = player_info['VTI'].strip()
94 subtitle = player_info.get('VSU', '').strip()
96 title += ' - %s' % subtitle
99 'id': player_info['VID'],
101 'description': player_info.get('VDE'),
102 'upload_date': unified_strdate(upload_date_str),
103 'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
105 qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
108 for format_id, format_dict in player_info['VSR'].items():
109 f = dict(format_dict)
110 versionCode = f.get('versionCode')
116 lang_rexs = [r'VO?%s' % langcode, r'VO?.-ST%s' % langcode]
118 None if versionCode is None else (
119 10 if any(re.match(r, versionCode) for r in lang_rexs)
122 if versionCode is not None:
123 # The original version with subtitles has lower relevance
124 if re.match(r'VO-ST(F|A)', versionCode):
126 # The version with sourds/mal subtitles has also lower relevance
127 elif re.match(r'VO?(F|A)-STM\1', versionCode):
130 'format_id': format_id,
131 'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
132 'language_preference': lang_pref,
133 'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
134 'width': int_or_none(f.get('width')),
135 'height': int_or_none(f.get('height')),
136 'tbr': int_or_none(f.get('bitrate')),
137 'quality': qfunc(f['quality']),
138 'source_preference': source_pref,
141 if f.get('mediaType') == 'rtmp':
142 format['url'] = f['streamer']
143 format['play_path'] = 'mp4:' + f['url']
144 format['ext'] = 'flv'
146 format['url'] = f['url']
148 formats.append(format)
150 self._sort_formats(formats)
152 info_dict['formats'] = formats
156 # It also uses the arte_vp_url url from the webpage to extract the information
157 class ArteTVCreativeIE(ArteTVPlus7IE):
158 IE_NAME = 'arte.tv:creative'
159 _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/(?:magazine?/)?(?P<id>[^?#]+)'
162 'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
166 'title': 'Folge 2 - Corporate Design',
167 'upload_date': '20131004',
170 'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
174 'title': 'Monty Python live (mostly)',
175 'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
176 'upload_date': '20140805',
181 class ArteTVFutureIE(ArteTVPlus7IE):
182 IE_NAME = 'arte.tv:future'
183 _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
186 'url': 'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
190 'title': 'Les champignons au secours de la planète',
191 'upload_date': '20131101',
195 def _real_extract(self, url):
196 anchor_id, lang = self._extract_url_info(url)
197 webpage = self._download_webpage(url, anchor_id)
198 row = get_element_by_id(anchor_id, webpage)
199 return self._extract_from_webpage(row, anchor_id, lang)
202 class ArteTVDDCIE(ArteTVPlus7IE):
203 IE_NAME = 'arte.tv:ddc'
204 _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
206 def _real_extract(self, url):
207 video_id, lang = self._extract_url_info(url)
210 elif lang == 'emission':
212 webpage = self._download_webpage(url, video_id)
213 scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
214 script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
215 javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
216 json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
217 return self._extract_from_json_url(json_url, video_id, lang)
220 class ArteTVConcertIE(ArteTVPlus7IE):
221 IE_NAME = 'arte.tv:concert'
222 _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)'
225 'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
226 'md5': '9ea035b7bd69696b67aa2ccaaa218161',
230 'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
231 'upload_date': '20140128',
232 'description': 'md5:486eb08f991552ade77439fe6d82c305',
237 class ArteTVEmbedIE(ArteTVPlus7IE):
238 IE_NAME = 'arte.tv:embed'
239 _VALID_URL = r'''(?x)
241 /playerv2/embed\.php\?json_url=
243 http://arte\.tv/papi/tvguide/videos/stream/player/
244 (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*
248 def _real_extract(self, url):
249 mobj = re.match(self._VALID_URL, url)
250 video_id = mobj.group('id')
251 lang = mobj.group('lang')
252 json_url = mobj.group('json_url')
253 return self._extract_from_json_url(json_url, video_id, lang)