1 from __future__ import unicode_literals
5 from .common import InfoExtractor
23 def _media_xml_tag(tag):
24 return '{http://search.yahoo.com/mrss/}%s' % tag
27 class MTVServicesInfoExtractor(InfoExtractor):
28 _MOBILE_TEMPLATE = None
32 def _id_from_uri(uri):
33 return uri.split(':')[-1]
35 # This was originally implemented for ComedyCentral, but it also works here
37 def _transform_rtmp_url(rtmp_video_url):
38 m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
41 base = 'http://viacommtvstrmfs.fplive.net/'
42 return base + m.group('finalid')
44 def _get_feed_url(self, uri):
47 def _get_thumbnail_url(self, uri, itemdoc):
48 search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
49 thumb_node = itemdoc.find(search_path)
50 if thumb_node is None:
53 return thumb_node.attrib['url']
55 def _extract_mobile_video_formats(self, mtvn_id):
56 webpage_url = self._MOBILE_TEMPLATE % mtvn_id
57 req = sanitized_Request(webpage_url)
58 # Otherwise we get a webpage that would execute some javascript
59 req.add_header('User-Agent', 'curl/7')
60 webpage = self._download_webpage(req, mtvn_id,
61 'Downloading mobile page')
62 metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
63 req = HEADRequest(metrics_url)
64 response = self._request_webpage(req, mtvn_id, 'Resolving url')
65 url = response.geturl()
66 # Transform the url to get the best quality:
67 url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
68 return [{'url': url, 'ext': 'mp4'}]
70 def _extract_video_formats(self, mdoc, mtvn_id):
71 if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc.find('.//src').text) is not None:
72 if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
73 self.to_screen('The normal version is not available from your '
74 'country, trying with the mobile version')
75 return self._extract_mobile_video_formats(mtvn_id)
76 raise ExtractorError('This video is not available from your country.',
80 for rendition in mdoc.findall('.//rendition'):
82 _, _, ext = rendition.attrib['type'].partition('/')
83 rtmp_video_url = rendition.find('./src').text
84 if rtmp_video_url.endswith('siteunavail.png'):
88 'url': self._transform_rtmp_url(rtmp_video_url),
89 'format_id': rendition.get('bitrate'),
90 'width': int(rendition.get('width')),
91 'height': int(rendition.get('height')),
93 except (KeyError, TypeError):
94 raise ExtractorError('Invalid rendition field.')
95 self._sort_formats(formats)
98 def _extract_subtitles(self, mdoc, mtvn_id):
100 for transcript in mdoc.findall('.//transcript'):
101 if transcript.get('kind') != 'captions':
103 lang = transcript.get('srclang')
105 'url': compat_str(typographic.get('src')),
106 'ext': typographic.get('format')
107 } for typographic in transcript.findall('./typographic')]
110 def _get_video_info(self, itemdoc):
111 uri = itemdoc.find('guid').text
112 video_id = self._id_from_uri(uri)
113 self.report_extraction(video_id)
114 mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
115 # Remove the templates, like &device={device}
116 mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
117 if 'acceptMethods' not in mediagen_url:
118 mediagen_url += '&' if '?' in mediagen_url else '?'
119 mediagen_url += 'acceptMethods=fms'
121 mediagen_doc = self._download_xml(mediagen_url, video_id,
122 'Downloading video urls')
124 item = mediagen_doc.find('./video/item')
125 if item is not None and item.get('type') == 'text':
126 message = '%s returned error: ' % self.IE_NAME
127 if item.get('code') is not None:
128 message += '%s - ' % item.get('code')
130 raise ExtractorError(message, expected=True)
132 description_node = itemdoc.find('description')
133 if description_node is not None:
134 description = description_node.text.strip()
140 title_el = find_xpath_attr(
141 itemdoc, './/{http://search.yahoo.com/mrss/}category',
142 'scheme', 'urn:mtvn:video_title')
144 title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
146 title_el = itemdoc.find('.//title') or itemdoc.find('./title')
147 if title_el.text is None:
150 title = title_el.text
152 raise ExtractorError('Could not find video title')
153 title = title.strip()
155 # This a short id that's used in the webpage urls
157 mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
158 'scheme', 'urn:mtvn:id')
159 if mtvn_id_node is not None:
160 mtvn_id = mtvn_id_node.text
162 content_el = find_xpath_attr(itemdoc, self._xpath_ns('.//content', 'http://search.yahoo.com/mrss/'), 'duration')
163 duration = float_or_none(content_el.attrib.get('duration')) if content_el is not None else None
167 'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
168 'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id),
170 'thumbnail': self._get_thumbnail_url(uri, itemdoc),
171 'description': description,
172 'duration': duration,
175 def _get_feed_query(self, uri):
178 data['lang'] = self._LANG
179 return compat_urllib_parse.urlencode(data)
181 def _get_videos_info(self, uri):
182 video_id = self._id_from_uri(uri)
183 feed_url = self._get_feed_url(uri)
184 info_url = feed_url + '?' + self._get_feed_query(uri)
185 return self._get_videos_info_from_url(info_url, video_id)
187 def _get_videos_info_from_url(self, url, video_id):
188 idoc = self._download_xml(
190 'Downloading info', transform_source=fix_xml_ampersands)
191 return self.playlist_result(
192 [self._get_video_info(item) for item in idoc.findall('.//item')])
194 def _extract_mgid(self, webpage):
196 # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
197 # or http://media.mtvnservices.com/{mgid}
198 og_url = self._og_search_video_url(webpage)
199 mgid = url_basename(og_url)
200 if mgid.endswith('.swf'):
202 except RegexNotFoundError:
205 if mgid is None or ':' not in mgid:
206 mgid = self._search_regex(
207 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
208 webpage, 'mgid', default=None)
211 sm4_embed = self._html_search_meta(
212 'sm4:video:embed', webpage, 'sm4 embed', default='')
213 mgid = self._search_regex(
214 r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid')
217 def _real_extract(self, url):
218 title = url_basename(url)
219 webpage = self._download_webpage(url, title)
220 mgid = self._extract_mgid(webpage)
221 videos_info = self._get_videos_info(mgid)
225 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
226 IE_NAME = 'mtvservices:embedded'
227 _VALID_URL = r'https?://media\.mtvnservices\.com/embed/(?P<mgid>.+?)(\?|/|$)'
230 # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
231 'url': 'http://media.mtvnservices.com/embed/mgid:uma:video:mtv.com:1043906/cp~vid%3D1043906%26uri%3Dmgid%3Auma%3Avideo%3Amtv.com%3A1043906',
232 'md5': 'cb349b21a7897164cede95bd7bf3fbb9',
236 'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds',
237 'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.',
242 def _extract_url(webpage):
244 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage)
246 return mobj.group('url')
248 def _get_feed_url(self, uri):
249 video_id = self._id_from_uri(uri)
250 site_id = uri.replace(video_id, '')
251 config_url = ('http://media.mtvnservices.com/pmt/e1/players/{0}/'
252 'context4/context5/config.xml'.format(site_id))
253 config_doc = self._download_xml(config_url, video_id)
254 feed_node = config_doc.find('.//feed')
255 feed_url = feed_node.text.strip().split('?')[0]
258 def _real_extract(self, url):
259 mobj = re.match(self._VALID_URL, url)
260 mgid = mobj.group('mgid')
261 return self._get_videos_info(mgid)
264 class MTVIE(MTVServicesInfoExtractor):
265 _VALID_URL = r'''(?x)^https?://
266 (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
267 m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
269 _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
273 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
274 'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
278 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
279 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
284 def _get_thumbnail_url(self, uri, itemdoc):
285 return 'http://mtv.mtvnimages.com/uri/' + uri
287 def _real_extract(self, url):
288 mobj = re.match(self._VALID_URL, url)
289 video_id = mobj.group('videoid')
290 uri = mobj.groupdict().get('mgid')
292 webpage = self._download_webpage(url, video_id)
294 # Some videos come from Vevo.com
296 r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage)
298 vevo_id = m_vevo.group(1)
299 self.to_screen('Vevo video detected: %s' % vevo_id)
300 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
302 uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
303 return self._get_videos_info(uri)
306 class MTVIggyIE(MTVServicesInfoExtractor):
307 IE_NAME = 'mtviggy.com'
308 _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+'
310 'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
314 'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
317 _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'
320 class MTVDEIE(MTVServicesInfoExtractor):
322 _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$'
324 'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
326 'id': 'music_video-a50bc5f0b3aa4b3190aa',
328 'title': 'MusicVideo_cro-traum',
329 'description': 'Cro - Traum',
333 'skip_download': True,
336 # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
337 'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen',
339 'id': 'local_playlist-f5ae778b9832cc837189',
341 'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
345 'skip_download': True,
348 # single video in pagePlaylist with different id
349 'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
351 'id': 'local_playlist-4e760566473c4c8c5344',
353 'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
354 'description': 'MTV Movies Supercut',
358 'skip_download': True,
362 def _real_extract(self, url):
363 video_id = self._match_id(url)
365 webpage = self._download_webpage(url, video_id)
367 playlist = self._parse_json(
369 r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'),
372 # news pages contain single video in playlist with different id
373 if len(playlist) == 1:
374 return self._get_videos_info_from_url(playlist[0]['mrss'], video_id)
376 for item in playlist:
377 item_id = item.get('id')
378 if item_id and compat_str(item_id) == video_id:
379 return self._get_videos_info_from_url(item['mrss'], video_id)