2 from __future__ import unicode_literals
6 from .common import InfoExtractor
14 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'],
44 'quality': 2 if q.attrib['quality'] == 'hd' else 1,
45 } for q in config.findall('./urls/url')]
46 self._sort_formats(formats)
48 title = config.find('.//name').text
49 thumbnail = config.find('.//firstThumbnailUrl').text
53 'thumbnail': thumbnail,
58 class ArteTVPlus7IE(InfoExtractor):
59 IE_NAME = 'arte.tv:+7'
60 _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
63 def _extract_url_info(cls, url):
64 mobj = re.match(cls._VALID_URL, url)
65 lang = mobj.group('lang')
66 # This is not a real id, it can be for example AJT for the news
67 # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
68 video_id = mobj.group('id')
71 def _real_extract(self, url):
72 video_id, lang = self._extract_url_info(url)
73 webpage = self._download_webpage(url, video_id)
74 return self._extract_from_webpage(webpage, video_id, lang)
76 def _extract_from_webpage(self, webpage, video_id, lang):
77 json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
78 return self._extract_from_json_url(json_url, video_id, lang)
80 def _extract_from_json_url(self, json_url, video_id, lang):
81 info = self._download_json(json_url, video_id)
82 player_info = info['videoJsonPlayer']
85 'id': player_info['VID'],
86 'title': player_info['VTI'],
87 'description': player_info.get('VDE'),
88 'upload_date': unified_strdate(player_info.get('VDA', '').split(' ')[0]),
89 'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
92 all_formats = player_info['VSR'].values()
93 # Some formats use the m3u8 protocol
94 all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats))
96 if f.get('versionCode') is None:
98 # Return true if that format is in the language of the url
105 regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
106 return any(re.match(r, f['versionCode']) for r in regexes)
107 # Some formats may not be in the same language as the url
108 formats = filter(_match_lang, all_formats)
109 formats = list(formats) # in python3 filter returns an iterator
111 # Some videos are only available in the 'Originalversion'
112 # they aren't tagged as being in French or German
113 if all(f['versionCode'] == 'VO' for f in all_formats):
114 formats = all_formats
116 raise ExtractorError(u'The formats list is empty')
118 if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
120 return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
124 # Sort first by quality
125 int(f.get('height',-1)),
126 int(f.get('bitrate',-1)),
127 # The original version with subtitles has lower relevance
128 re.match(r'VO-ST(F|A)', f.get('versionCode', '')) is None,
129 # The version with sourds/mal subtitles has also lower relevance
130 re.match(r'VO?(F|A)-STM\1', f.get('versionCode', '')) is None,
131 # Prefer http downloads over m3u8
132 0 if f['url'].endswith('m3u8') else 1,
134 formats = sorted(formats, key=sort_key)
135 def _format(format_info):
137 height = format_info.get('height')
138 if height is not None:
139 quality = compat_str(height)
140 bitrate = format_info.get('bitrate')
141 if bitrate is not None:
142 quality += '-%d' % bitrate
143 if format_info.get('versionCode') is not None:
144 format_id = '%s-%s' % (quality, format_info['versionCode'])
148 'format_id': format_id,
149 'format_note': format_info.get('versionLibelle'),
150 'width': format_info.get('width'),
153 if format_info['mediaType'] == 'rtmp':
154 info['url'] = format_info['streamer']
155 info['play_path'] = 'mp4:' + format_info['url']
158 info['url'] = format_info['url']
159 info['ext'] = determine_ext(info['url'])
161 info_dict['formats'] = [_format(f) for f in formats]
166 # It also uses the arte_vp_url url from the webpage to extract the information
167 class ArteTVCreativeIE(ArteTVPlus7IE):
168 IE_NAME = 'arte.tv:creative'
169 _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/magazine?/(?P<id>.+)'
172 'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
176 'title': 'Agentur Amateur / Agence Amateur #2 : Corporate Design',
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',
194 def _real_extract(self, url):
195 anchor_id, lang = self._extract_url_info(url)
196 webpage = self._download_webpage(url, anchor_id)
197 row = get_element_by_id(anchor_id, webpage)
198 return self._extract_from_webpage(row, anchor_id, lang)
201 class ArteTVDDCIE(ArteTVPlus7IE):
202 IE_NAME = 'arte.tv:ddc'
203 _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
205 def _real_extract(self, url):
206 video_id, lang = self._extract_url_info(url)
209 elif lang == 'emission':
211 webpage = self._download_webpage(url, video_id)
212 scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
213 script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
214 javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
215 json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
216 return self._extract_from_json_url(json_url, video_id, lang)
219 class ArteTVConcertIE(ArteTVPlus7IE):
220 IE_NAME = 'arte.tv:concert'
221 _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)'
224 'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
225 'md5': '9ea035b7bd69696b67aa2ccaaa218161',
229 'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
230 'upload_date': '20140128',
231 'description': 'md5:486eb08f991552ade77439fe6d82c305',
236 class ArteTVEmbedIE(ArteTVPlus7IE):
237 IE_NAME = 'arte.tv:embed'
238 _VALID_URL = r'''(?x)
240 /playerv2/embed\.php\?json_url=
242 http://arte\.tv/papi/tvguide/videos/stream/player/
243 (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*
247 def _real_extract(self, url):
248 mobj = re.match(self._VALID_URL, url)
249 video_id = mobj.group('id')
250 lang = mobj.group('lang')
251 json_url = mobj.group('json_url')
252 return self._extract_from_json_url(json_url, video_id, lang)