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
27 def _id_from_uri(uri
):
28 return uri
.split(':')[-1]
30 # This was originally implemented for ComedyCentral, but it also works here
32 def _transform_rtmp_url(rtmp_video_url
):
33 m
= re
.match(r
'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url
)
36 base
= 'http://viacommtvstrmfs.fplive.net/'
37 return base
+ m
.group('finalid')
39 def _get_feed_url(self
, uri
):
42 def _get_thumbnail_url(self
, uri
, itemdoc
):
43 search_path
= '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
44 thumb_node
= itemdoc
.find(search_path
)
45 if thumb_node
is None:
48 return thumb_node
.attrib
['url']
50 def _extract_mobile_video_formats(self
, mtvn_id
):
51 webpage_url
= self
._MOBILE
_TEMPLATE
% mtvn_id
52 req
= compat_urllib_request
.Request(webpage_url
)
53 # Otherwise we get a webpage that would execute some javascript
54 req
.add_header('Youtubedl-user-agent', 'curl/7')
55 webpage
= self
._download
_webpage
(req
, mtvn_id
,
56 'Downloading mobile page')
57 metrics_url
= unescapeHTML(self
._search
_regex
(r
'<a href="(http://metrics.+?)"', webpage
, 'url'))
58 req
= HEADRequest(metrics_url
)
59 response
= self
._request
_webpage
(req
, mtvn_id
, 'Resolving url')
60 url
= response
.geturl()
61 # Transform the url to get the best quality:
62 url
= re
.sub(r
'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url
, 1)
63 return [{'url': url,'ext': 'mp4'}
]
65 def _extract_video_formats(self
, mdoc
, mtvn_id
):
66 if re
.match(r
'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc
.find('.//src').text
) is not None:
67 if mtvn_id
is not None and self
._MOBILE
_TEMPLATE
is not None:
68 self
.to_screen('The normal version is not available from your '
69 'country, trying with the mobile version')
70 return self
._extract
_mobile
_video
_formats
(mtvn_id
)
71 raise ExtractorError('This video is not available from your country.',
75 for rendition
in mdoc
.findall('.//rendition'):
77 _
, _
, ext
= rendition
.attrib
['type'].partition('/')
78 rtmp_video_url
= rendition
.find('./src').text
79 formats
.append({'ext': ext
,
80 'url': self
._transform
_rtmp
_url
(rtmp_video_url
),
81 'format_id': rendition
.get('bitrate'),
82 'width': int(rendition
.get('width')),
83 'height': int(rendition
.get('height')),
85 except (KeyError, TypeError):
86 raise ExtractorError('Invalid rendition field.')
87 self
._sort
_formats
(formats
)
90 def _get_video_info(self
, itemdoc
):
91 uri
= itemdoc
.find('guid').text
92 video_id
= self
._id
_from
_uri
(uri
)
93 self
.report_extraction(video_id
)
94 mediagen_url
= itemdoc
.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib
['url']
95 # Remove the templates, like &device={device}
96 mediagen_url
= re
.sub(r
'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url
)
97 if 'acceptMethods' not in mediagen_url
:
98 mediagen_url
+= '&acceptMethods=fms'
100 mediagen_doc
= self
._download
_xml
(mediagen_url
, video_id
,
101 'Downloading video urls')
103 description_node
= itemdoc
.find('description')
104 if description_node
is not None:
105 description
= description_node
.text
.strip()
111 title_el
= find_xpath_attr(
112 itemdoc
, './/{http://search.yahoo.com/mrss/}category',
113 'scheme', 'urn:mtvn:video_title')
115 title_el
= itemdoc
.find('.//{http://search.yahoo.com/mrss/}title')
117 title_el
= itemdoc
.find('.//title')
118 if title_el
.text
is None:
121 title
= title_el
.text
123 raise ExtractorError('Could not find video title')
124 title
= title
.strip()
126 # This a short id that's used in the webpage urls
128 mtvn_id_node
= find_xpath_attr(itemdoc
, './/{http://search.yahoo.com/mrss/}category',
129 'scheme', 'urn:mtvn:id')
130 if mtvn_id_node
is not None:
131 mtvn_id
= mtvn_id_node
.text
135 'formats': self
._extract
_video
_formats
(mediagen_doc
, mtvn_id
),
137 'thumbnail': self
._get
_thumbnail
_url
(uri
, itemdoc
),
138 'description': description
,
141 def _get_videos_info(self
, uri
):
142 video_id
= self
._id
_from
_uri
(uri
)
143 feed_url
= self
._get
_feed
_url
(uri
)
144 data
= compat_urllib_parse
.urlencode({'uri': uri}
)
145 idoc
= self
._download
_xml
(
146 feed_url
+ '?' + data
, video_id
,
147 'Downloading info', transform_source
=fix_xml_ampersands
)
148 return [self
._get
_video
_info
(item
) for item
in idoc
.findall('.//item')]
150 def _real_extract(self
, url
):
151 title
= url_basename(url
)
152 webpage
= self
._download
_webpage
(url
, title
)
154 # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
155 # or http://media.mtvnservices.com/{mgid}
156 og_url
= self
._og
_search
_video
_url
(webpage
)
157 mgid
= url_basename(og_url
)
158 if mgid
.endswith('.swf'):
160 except RegexNotFoundError
:
163 if mgid
is None or ':' not in mgid
:
164 mgid
= self
._search
_regex
(
165 [r
'data-mgid="(.*?)"', r
'swfobject.embedSWF\(".*?(mgid:.*?)"'],
167 return self
._get
_videos
_info
(mgid
)
170 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor
):
171 IE_NAME
= 'mtvservices:embedded'
172 _VALID_URL
= r
'https?://media\.mtvnservices\.com/embed/(?P<mgid>.+?)(\?|/|$)'
175 # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
176 'url': 'http://media.mtvnservices.com/embed/mgid:uma:video:mtv.com:1043906/cp~vid%3D1043906%26uri%3Dmgid%3Auma%3Avideo%3Amtv.com%3A1043906',
177 'md5': 'cb349b21a7897164cede95bd7bf3fbb9',
181 'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds',
182 'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.',
186 def _get_feed_url(self
, uri
):
187 video_id
= self
._id
_from
_uri
(uri
)
188 site_id
= uri
.replace(video_id
, '')
189 config_url
= 'http://media.mtvnservices.com/pmt/e1/players/{0}/config.xml'.format(site_id
)
190 config_doc
= self
._download
_xml
(config_url
, video_id
)
191 feed_node
= config_doc
.find('.//feed')
192 feed_url
= feed_node
.text
.strip().split('?')[0]
195 def _real_extract(self
, url
):
196 mobj
= re
.match(self
._VALID
_URL
, url
)
197 mgid
= mobj
.group('mgid')
198 return self
._get
_videos
_info
(mgid
)
201 class MTVIE(MTVServicesInfoExtractor
):
202 _VALID_URL
= r
'''(?x)^https?://
203 (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
204 m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
206 _FEED_URL
= 'http://www.mtv.com/player/embed/AS3/rss/'
210 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
211 'file': '853555.mp4',
212 'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
214 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
215 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
220 'url': 'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
221 'file': 'USCJY1331283.mp4',
222 'md5': '73b4e7fcadd88929292fe52c3ced8caf',
224 'title': 'Everything Has Changed',
225 'upload_date': '20130606',
226 'uploader': 'Taylor Swift',
228 'skip': 'VEVO is only available in some countries',
232 def _get_thumbnail_url(self
, uri
, itemdoc
):
233 return 'http://mtv.mtvnimages.com/uri/' + uri
235 def _real_extract(self
, url
):
236 mobj
= re
.match(self
._VALID
_URL
, url
)
237 video_id
= mobj
.group('videoid')
238 uri
= mobj
.groupdict().get('mgid')
240 webpage
= self
._download
_webpage
(url
, video_id
)
242 # Some videos come from Vevo.com
243 m_vevo
= re
.search(r
'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
246 vevo_id
= m_vevo
.group(1);
247 self
.to_screen('Vevo video detected: %s' % vevo_id
)
248 return self
.url_result('vevo:%s' % vevo_id
, ie
='Vevo')
250 uri
= self
._html
_search
_regex
(r
'/uri/(.*?)\?', webpage
, 'uri')
251 return self
._get
_videos
_info
(uri
)
254 class MTVIggyIE(MTVServicesInfoExtractor
):
255 IE_NAME
= 'mtviggy.com'
256 _VALID_URL
= r
'https?://www\.mtviggy\.com/videos/.+'
258 'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
262 'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
265 _FEED_URL
= 'http://all.mtvworldverticals.com/feed-xml/'