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
31 def _id_from_uri(uri
):
32 return uri
.split(':')[-1]
34 # This was originally implemented for ComedyCentral, but it also works here
36 def _transform_rtmp_url(rtmp_video_url
):
37 m
= re
.match(r
'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url
)
40 base
= 'http://viacommtvstrmfs.fplive.net/'
41 return base
+ m
.group('finalid')
43 def _get_feed_url(self
, uri
):
46 def _get_thumbnail_url(self
, uri
, itemdoc
):
47 search_path
= '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
48 thumb_node
= itemdoc
.find(search_path
)
49 if thumb_node
is None:
52 return thumb_node
.attrib
['url']
54 def _extract_mobile_video_formats(self
, mtvn_id
):
55 webpage_url
= self
._MOBILE
_TEMPLATE
% mtvn_id
56 req
= sanitized_Request(webpage_url
)
57 # Otherwise we get a webpage that would execute some javascript
58 req
.add_header('User-Agent', 'curl/7')
59 webpage
= self
._download
_webpage
(req
, mtvn_id
,
60 'Downloading mobile page')
61 metrics_url
= unescapeHTML(self
._search
_regex
(r
'<a href="(http://metrics.+?)"', webpage
, 'url'))
62 req
= HEADRequest(metrics_url
)
63 response
= self
._request
_webpage
(req
, mtvn_id
, 'Resolving url')
64 url
= response
.geturl()
65 # Transform the url to get the best quality:
66 url
= re
.sub(r
'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url
, 1)
67 return [{'url': url, 'ext': 'mp4'}
]
69 def _extract_video_formats(self
, mdoc
, mtvn_id
):
70 if re
.match(r
'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc
.find('.//src').text
) is not None:
71 if mtvn_id
is not None and self
._MOBILE
_TEMPLATE
is not None:
72 self
.to_screen('The normal version is not available from your '
73 'country, trying with the mobile version')
74 return self
._extract
_mobile
_video
_formats
(mtvn_id
)
75 raise ExtractorError('This video is not available from your country.',
79 for rendition
in mdoc
.findall('.//rendition'):
81 _
, _
, ext
= rendition
.attrib
['type'].partition('/')
82 rtmp_video_url
= rendition
.find('./src').text
83 if rtmp_video_url
.endswith('siteunavail.png'):
87 'url': self
._transform
_rtmp
_url
(rtmp_video_url
),
88 'format_id': rendition
.get('bitrate'),
89 'width': int(rendition
.get('width')),
90 'height': int(rendition
.get('height')),
92 except (KeyError, TypeError):
93 raise ExtractorError('Invalid rendition field.')
94 self
._sort
_formats
(formats
)
97 def _extract_subtitles(self
, mdoc
, mtvn_id
):
99 for transcript
in mdoc
.findall('.//transcript'):
100 if transcript
.get('kind') != 'captions':
102 lang
= transcript
.get('srclang')
104 'url': compat_str(typographic
.get('src')),
105 'ext': typographic
.get('format')
106 } for typographic
in transcript
.findall('./typographic')]
109 def _get_video_info(self
, itemdoc
):
110 uri
= itemdoc
.find('guid').text
111 video_id
= self
._id
_from
_uri
(uri
)
112 self
.report_extraction(video_id
)
113 mediagen_url
= itemdoc
.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib
['url']
114 # Remove the templates, like &device={device}
115 mediagen_url
= re
.sub(r
'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url
)
116 if 'acceptMethods' not in mediagen_url
:
117 mediagen_url
+= '&' if '?' in mediagen_url
else '?'
118 mediagen_url
+= 'acceptMethods=fms'
120 mediagen_doc
= self
._download
_xml
(mediagen_url
, video_id
,
121 'Downloading video urls')
123 item
= mediagen_doc
.find('./video/item')
124 if item
is not None and item
.get('type') == 'text':
125 message
= '%s returned error: ' % self
.IE_NAME
126 if item
.get('code') is not None:
127 message
+= '%s - ' % item
.get('code')
129 raise ExtractorError(message
, expected
=True)
131 description_node
= itemdoc
.find('description')
132 if description_node
is not None:
133 description
= description_node
.text
.strip()
139 title_el
= find_xpath_attr(
140 itemdoc
, './/{http://search.yahoo.com/mrss/}category',
141 'scheme', 'urn:mtvn:video_title')
143 title_el
= itemdoc
.find('.//{http://search.yahoo.com/mrss/}title')
145 title_el
= itemdoc
.find('.//title') or itemdoc
.find('./title')
146 if title_el
.text
is None:
149 title
= title_el
.text
151 raise ExtractorError('Could not find video title')
152 title
= title
.strip()
154 # This a short id that's used in the webpage urls
156 mtvn_id_node
= find_xpath_attr(itemdoc
, './/{http://search.yahoo.com/mrss/}category',
157 'scheme', 'urn:mtvn:id')
158 if mtvn_id_node
is not None:
159 mtvn_id
= mtvn_id_node
.text
163 'formats': self
._extract
_video
_formats
(mediagen_doc
, mtvn_id
),
164 'subtitles': self
._extract
_subtitles
(mediagen_doc
, mtvn_id
),
166 'thumbnail': self
._get
_thumbnail
_url
(uri
, itemdoc
),
167 'description': description
,
170 def _get_feed_query(self
, uri
):
173 data
['lang'] = self
._LANG
174 return compat_urllib_parse
.urlencode(data
)
176 def _get_videos_info(self
, uri
):
177 video_id
= self
._id
_from
_uri
(uri
)
178 feed_url
= self
._get
_feed
_url
(uri
)
179 info_url
= feed_url
+ '?' + self
._get
_feed
_query
(uri
)
180 return self
._get
_videos
_info
_from
_url
(info_url
, video_id
)
182 def _get_videos_info_from_url(self
, url
, video_id
):
183 idoc
= self
._download
_xml
(
185 'Downloading info', transform_source
=fix_xml_ampersands
)
186 return self
.playlist_result(
187 [self
._get
_video
_info
(item
) for item
in idoc
.findall('.//item')])
189 def _extract_mgid(self
, webpage
):
191 # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
192 # or http://media.mtvnservices.com/{mgid}
193 og_url
= self
._og
_search
_video
_url
(webpage
)
194 mgid
= url_basename(og_url
)
195 if mgid
.endswith('.swf'):
197 except RegexNotFoundError
:
200 if mgid
is None or ':' not in mgid
:
201 mgid
= self
._search
_regex
(
202 [r
'data-mgid="(.*?)"', r
'swfobject.embedSWF\(".*?(mgid:.*?)"'],
203 webpage
, 'mgid', default
=None)
206 sm4_embed
= self
._html
_search
_meta
(
207 'sm4:video:embed', webpage
, 'sm4 embed', default
='')
208 mgid
= self
._search
_regex
(
209 r
'embed/(mgid:.+?)["\'&?
/]', sm4_embed, 'mgid
')
212 def _real_extract(self, url):
213 title = url_basename(url)
214 webpage = self._download_webpage(url, title)
215 mgid = self._extract_mgid(webpage)
216 videos_info = self._get_videos_info(mgid)
220 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
221 IE_NAME = 'mtvservices
:embedded
'
222 _VALID_URL = r'https?
://media\
.mtvnservices\
.com
/embed
/(?P
<mgid
>.+?
)(\?|
/|$
)'
225 # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
226 'url
': 'http
://media
.mtvnservices
.com
/embed
/mgid
:uma
:video
:mtv
.com
:1043906/cp~vid
%3D1043906
%26uri
%3Dmgid
%3Auma
%3Avideo
%3Amtv
.com
%3A1043906
',
227 'md5
': 'cb349b21a7897164cede95bd7bf3fbb9
',
231 'title
': 'Peter Dinklage Sums Up
\'Game Of Thrones
\' In
45 Seconds
',
232 'description
': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage
as he tries summarizing
"Game of Thrones" in under a minute
.',
237 def _extract_url(webpage):
239 r'<iframe
[^
>]+?src
=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage)
241 return mobj.group('url')
243 def _get_feed_url(self, uri):
244 video_id = self._id_from_uri(uri)
245 site_id = uri.replace(video_id, '')
246 config_url = ('http://media.mtvnservices.com/pmt/e1/players/{0}/'
247 'context4/context5/config.xml'.format(site_id))
248 config_doc = self._download_xml(config_url, video_id)
249 feed_node = config_doc.find('.//feed')
250 feed_url = feed_node.text.strip().split('?')[0]
253 def _real_extract(self, url):
254 mobj = re.match(self._VALID_URL, url)
255 mgid = mobj.group('mgid')
256 return self._get_videos_info(mgid)
259 class MTVIE(MTVServicesInfoExtractor):
260 _VALID_URL = r'''(?x)^https?://
261 (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
262 m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
264 _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
268 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
269 'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
273 'title': 'Taylor Swift - "Ours (VH1 Storytellers
)"',
274 'description': 'Album: Taylor Swift performs "Ours
" for VH1 Storytellers at Harvey Mudd College.',
279 def _get_thumbnail_url(self, uri, itemdoc):
280 return 'http://mtv.mtvnimages.com/uri/' + uri
282 def _real_extract(self, url):
283 mobj = re.match(self._VALID_URL, url)
284 video_id = mobj.group('videoid')
285 uri = mobj.groupdict().get('mgid')
287 webpage = self._download_webpage(url, video_id)
289 # Some videos come from Vevo.com
291 r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?
)";', webpage)
293 vevo_id = m_vevo.group(1)
294 self.to_screen('Vevo video detected: %s' % vevo_id)
295 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
297 uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
298 return self._get_videos_info(uri)
301 class MTVIggyIE(MTVServicesInfoExtractor):
302 IE_NAME = 'mtviggy.com'
303 _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+'
305 'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
309 'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
312 _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'
315 class MTVDEIE(MTVServicesInfoExtractor):
317 _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$'
319 'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
321 'id': 'music_video-a50bc5f0b3aa4b3190aa',
323 'title': 'MusicVideo_cro-traum',
324 'description': 'Cro - Traum',
328 'skip_download': True,
331 # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
332 'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen',
334 'id': 'local_playlist-f5ae778b9832cc837189',
336 'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
340 'skip_download': True,
343 # single video in pagePlaylist with different id
344 'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
346 'id': 'local_playlist-4e760566473c4c8c5344',
348 'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
349 'description': 'MTV Movies Supercut',
353 'skip_download': True,
357 def _real_extract(self, url):
358 video_id = self._match_id(url)
360 webpage = self._download_webpage(url, video_id)
362 playlist = self._parse_json(
364 r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'),
367 # news pages contain single video in playlist with different id
368 if len(playlist) == 1:
369 return self._get_videos_info_from_url(playlist[0]['mrss'], video_id)
371 for item in playlist:
372 item_id = item.get('id')
373 if item_id and compat_str(item_id) == video_id:
374 return self._get_videos_info_from_url(item['mrss'], video_id)