1 from __future__ import unicode_literals
5 from .common import InfoExtractor
22 def _media_xml_tag(tag):
23 return '{http://search.yahoo.com/mrss/}%s' % tag
26 class MTVServicesInfoExtractor(InfoExtractor):
27 _MOBILE_TEMPLATE = None
30 def _id_from_uri(uri):
31 return uri.split(':')[-1]
33 # This was originally implemented for ComedyCentral, but it also works here
35 def _transform_rtmp_url(rtmp_video_url):
36 m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
39 base = 'http://viacommtvstrmfs.fplive.net/'
40 return base + m.group('finalid')
42 def _get_feed_url(self, uri):
45 def _get_thumbnail_url(self, uri, itemdoc):
46 search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
47 thumb_node = itemdoc.find(search_path)
48 if thumb_node is None:
51 return thumb_node.attrib['url']
53 def _extract_mobile_video_formats(self, mtvn_id):
54 webpage_url = self._MOBILE_TEMPLATE % mtvn_id
55 req = compat_urllib_request.Request(webpage_url)
56 # Otherwise we get a webpage that would execute some javascript
57 req.add_header('User-Agent', 'curl/7')
58 webpage = self._download_webpage(req, mtvn_id,
59 'Downloading mobile page')
60 metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
61 req = HEADRequest(metrics_url)
62 response = self._request_webpage(req, mtvn_id, 'Resolving url')
63 url = response.geturl()
64 # Transform the url to get the best quality:
65 url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
66 return [{'url': url, 'ext': 'mp4'}]
68 def _extract_video_formats(self, mdoc, mtvn_id):
69 if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None:
70 if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
71 self.to_screen('The normal version is not available from your '
72 'country, trying with the mobile version')
73 return self._extract_mobile_video_formats(mtvn_id)
74 raise ExtractorError('This video is not available from your country.',
78 for rendition in mdoc.findall('.//rendition'):
80 _, _, ext = rendition.attrib['type'].partition('/')
81 rtmp_video_url = rendition.find('./src').text
82 if rtmp_video_url.endswith('siteunavail.png'):
86 'url': self._transform_rtmp_url(rtmp_video_url),
87 'format_id': rendition.get('bitrate'),
88 'width': int(rendition.get('width')),
89 'height': int(rendition.get('height')),
91 except (KeyError, TypeError):
92 raise ExtractorError('Invalid rendition field.')
93 self._sort_formats(formats)
96 def _extract_subtitles(self, mdoc, mtvn_id):
98 for transcript in mdoc.findall('.//transcript'):
99 if transcript.get('kind') != 'captions':
101 lang = transcript.get('srclang')
103 'url': compat_str(typographic.get('src')),
104 'ext': typographic.get('format')
105 } for typographic in transcript.findall('./typographic')]
108 def _get_video_info(self, itemdoc):
109 uri = itemdoc.find('guid').text
110 video_id = self._id_from_uri(uri)
111 self.report_extraction(video_id)
112 mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
113 # Remove the templates, like &device={device}
114 mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
115 if 'acceptMethods' not in mediagen_url:
116 mediagen_url += '&acceptMethods=fms'
118 mediagen_doc = self._download_xml(mediagen_url, video_id,
119 'Downloading video urls')
121 description_node = itemdoc.find('description')
122 if description_node is not None:
123 description = description_node.text.strip()
129 title_el = find_xpath_attr(
130 itemdoc, './/{http://search.yahoo.com/mrss/}category',
131 'scheme', 'urn:mtvn:video_title')
133 title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
135 title_el = itemdoc.find('.//title')
136 if title_el.text is None:
139 title = title_el.text
141 raise ExtractorError('Could not find video title')
142 title = title.strip()
144 # This a short id that's used in the webpage urls
146 mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
147 'scheme', 'urn:mtvn:id')
148 if mtvn_id_node is not None:
149 mtvn_id = mtvn_id_node.text
153 'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
154 'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id),
156 'thumbnail': self._get_thumbnail_url(uri, itemdoc),
157 'description': description,
160 def _get_videos_info(self, uri):
161 video_id = self._id_from_uri(uri)
162 feed_url = self._get_feed_url(uri)
163 data = compat_urllib_parse.urlencode({'uri': uri})
164 idoc = self._download_xml(
165 feed_url + '?' + data, video_id,
166 'Downloading info', transform_source=fix_xml_ampersands)
167 return self.playlist_result(
168 [self._get_video_info(item) for item in idoc.findall('.//item')])
170 def _real_extract(self, url):
171 title = url_basename(url)
172 webpage = self._download_webpage(url, title)
174 # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
175 # or http://media.mtvnservices.com/{mgid}
176 og_url = self._og_search_video_url(webpage)
177 mgid = url_basename(og_url)
178 if mgid.endswith('.swf'):
180 except RegexNotFoundError:
183 if mgid is None or ':' not in mgid:
184 mgid = self._search_regex(
185 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
188 videos_info = self._get_videos_info(mgid)
192 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
193 IE_NAME = 'mtvservices:embedded'
194 _VALID_URL = r'https?://media\.mtvnservices\.com/embed/(?P<mgid>.+?)(\?|/|$)'
197 # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
198 'url': 'http://media.mtvnservices.com/embed/mgid:uma:video:mtv.com:1043906/cp~vid%3D1043906%26uri%3Dmgid%3Auma%3Avideo%3Amtv.com%3A1043906',
199 'md5': 'cb349b21a7897164cede95bd7bf3fbb9',
203 'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds',
204 'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.',
208 def _get_feed_url(self, uri):
209 video_id = self._id_from_uri(uri)
210 site_id = uri.replace(video_id, '')
211 config_url = ('http://media.mtvnservices.com/pmt/e1/players/{0}/'
212 'context4/context5/config.xml'.format(site_id))
213 config_doc = self._download_xml(config_url, video_id)
214 feed_node = config_doc.find('.//feed')
215 feed_url = feed_node.text.strip().split('?')[0]
218 def _real_extract(self, url):
219 mobj = re.match(self._VALID_URL, url)
220 mgid = mobj.group('mgid')
221 return self._get_videos_info(mgid)
224 class MTVIE(MTVServicesInfoExtractor):
225 _VALID_URL = r'''(?x)^https?://
226 (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
227 m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
229 _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
233 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
234 'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
238 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
239 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
244 def _get_thumbnail_url(self, uri, itemdoc):
245 return 'http://mtv.mtvnimages.com/uri/' + uri
247 def _real_extract(self, url):
248 mobj = re.match(self._VALID_URL, url)
249 video_id = mobj.group('videoid')
250 uri = mobj.groupdict().get('mgid')
252 webpage = self._download_webpage(url, video_id)
254 # Some videos come from Vevo.com
256 r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage)
258 vevo_id = m_vevo.group(1)
259 self.to_screen('Vevo video detected: %s' % vevo_id)
260 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
262 uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
263 return self._get_videos_info(uri)
266 class MTVIggyIE(MTVServicesInfoExtractor):
267 IE_NAME = 'mtviggy.com'
268 _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+'
270 'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
274 'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
277 _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'