1 from __future__ import unicode_literals
5 from .common import InfoExtractor
19 def _media_xml_tag(tag):
20 return '{http://search.yahoo.com/mrss/}%s' % tag
23 class MTVServicesInfoExtractor(InfoExtractor):
24 _MOBILE_TEMPLATE = None
26 def _id_from_uri(uri):
27 return uri.split(':')[-1]
29 # This was originally implemented for ComedyCentral, but it also works here
31 def _transform_rtmp_url(rtmp_video_url):
32 m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
35 base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
36 return base + m.group('finalid')
38 def _get_thumbnail_url(self, uri, itemdoc):
39 search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
40 thumb_node = itemdoc.find(search_path)
41 if thumb_node is None:
44 return thumb_node.attrib['url']
46 def _extract_mobile_video_formats(self, mtvn_id):
47 webpage_url = self._MOBILE_TEMPLATE % mtvn_id
48 req = compat_urllib_request.Request(webpage_url)
49 # Otherwise we get a webpage that would execute some javascript
50 req.add_header('Youtubedl-user-agent', 'curl/7')
51 webpage = self._download_webpage(req, mtvn_id,
52 'Downloading mobile page')
53 metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
54 req = HEADRequest(metrics_url)
55 response = self._request_webpage(req, mtvn_id, 'Resolving url')
56 url = response.geturl()
57 # Transform the url to get the best quality:
58 url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
59 return [{'url': url,'ext': 'mp4'}]
61 def _extract_video_formats(self, mdoc, mtvn_id):
62 if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None:
63 if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
64 self.to_screen('The normal version is not available from your '
65 'country, trying with the mobile version')
66 return self._extract_mobile_video_formats(mtvn_id)
67 raise ExtractorError('This video is not available from your country.',
71 for rendition in mdoc.findall('.//rendition'):
73 _, _, ext = rendition.attrib['type'].partition('/')
74 rtmp_video_url = rendition.find('./src').text
75 formats.append({'ext': ext,
76 'url': self._transform_rtmp_url(rtmp_video_url),
77 'format_id': rendition.get('bitrate'),
78 'width': int(rendition.get('width')),
79 'height': int(rendition.get('height')),
81 except (KeyError, TypeError):
82 raise ExtractorError('Invalid rendition field.')
83 # worst format is expected to be first and best one last
84 formats.sort(key=lambda x: int(x['format_id']))
87 def _get_video_info(self, itemdoc):
88 uri = itemdoc.find('guid').text
89 video_id = self._id_from_uri(uri)
90 self.report_extraction(video_id)
91 mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
92 # Remove the templates, like &device={device}
93 mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
94 if 'acceptMethods' not in mediagen_url:
95 mediagen_url += '&acceptMethods=fms'
97 mediagen_doc = self._download_xml(mediagen_url, video_id,
98 'Downloading video urls')
100 description_node = itemdoc.find('description')
101 if description_node is not None:
102 description = description_node.text.strip()
108 title_el = find_xpath_attr(
109 itemdoc, './/{http://search.yahoo.com/mrss/}category',
110 'scheme', 'urn:mtvn:video_title')
112 title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
114 title_el = itemdoc.find('.//title')
115 if title_el.text is None:
118 title = title_el.text
120 raise ExtractorError('Could not find video title')
121 title = title.strip()
123 # This a short id that's used in the webpage urls
125 mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
126 'scheme', 'urn:mtvn:id')
127 if mtvn_id_node is not None:
128 mtvn_id = mtvn_id_node.text
132 'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
134 'thumbnail': self._get_thumbnail_url(uri, itemdoc),
135 'description': description,
138 def _get_videos_info(self, uri):
139 video_id = self._id_from_uri(uri)
140 data = compat_urllib_parse.urlencode({'uri': uri})
142 idoc = self._download_xml(
143 self._FEED_URL + '?' + data, video_id,
144 'Downloading info', transform_source=fix_xml_ampersands)
145 return [self._get_video_info(item) for item in idoc.findall('.//item')]
147 def _real_extract(self, url):
148 title = url_basename(url)
149 webpage = self._download_webpage(url, title)
151 # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
152 # or http://media.mtvnservices.com/{mgid}
153 og_url = self._og_search_video_url(webpage)
154 mgid = url_basename(og_url)
155 if mgid.endswith('.swf'):
157 except RegexNotFoundError:
158 mgid = self._search_regex(
159 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
161 return self._get_videos_info(mgid)
164 class MTVIE(MTVServicesInfoExtractor):
165 _VALID_URL = r'''(?x)^https?://
166 (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
167 m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
169 _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
173 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
174 'file': '853555.mp4',
175 'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
177 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
178 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
183 'url': 'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
184 'file': 'USCJY1331283.mp4',
185 'md5': '73b4e7fcadd88929292fe52c3ced8caf',
187 'title': 'Everything Has Changed',
188 'upload_date': '20130606',
189 'uploader': 'Taylor Swift',
191 'skip': 'VEVO is only available in some countries',
195 def _get_thumbnail_url(self, uri, itemdoc):
196 return 'http://mtv.mtvnimages.com/uri/' + uri
198 def _real_extract(self, url):
199 mobj = re.match(self._VALID_URL, url)
200 video_id = mobj.group('videoid')
201 uri = mobj.groupdict().get('mgid')
203 webpage = self._download_webpage(url, video_id)
205 # Some videos come from Vevo.com
206 m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
209 vevo_id = m_vevo.group(1);
210 self.to_screen('Vevo video detected: %s' % vevo_id)
211 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
213 uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
214 return self._get_videos_info(uri)
217 class MTVIggyIE(MTVServicesInfoExtractor):
218 IE_NAME = 'mtviggy.com'
219 _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+'
221 'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
225 'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
228 _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'