1 from __future__
import unicode_literals
5 from .common
import InfoExtractor
27 def _media_xml_tag(tag
):
28 return '{http://search.yahoo.com/mrss/}%s' % tag
31 class MTVServicesInfoExtractor(InfoExtractor
):
32 _MOBILE_TEMPLATE
= None
36 def _id_from_uri(uri
):
37 return uri
.split(':')[-1]
40 def _remove_template_parameter(url
):
41 # Remove the templates, like &device={device}
42 return re
.sub(r
'&[^=]*?={.*?}(?=(&|$))', '', url
)
44 # This was originally implemented for ComedyCentral, but it also works here
46 def _transform_rtmp_url(cls
, rtmp_video_url
):
47 m
= re
.match(r
'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url
)
49 return {'rtmp': rtmp_video_url}
50 base
= 'http://viacommtvstrmfs.fplive.net/'
51 return {'http': base + m.group('finalid')}
53 def _get_feed_url(self
, uri
):
56 def _get_thumbnail_url(self
, uri
, itemdoc
):
57 search_path
= '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
58 thumb_node
= itemdoc
.find(search_path
)
59 if thumb_node
is None:
62 return thumb_node
.attrib
['url']
64 def _extract_mobile_video_formats(self
, mtvn_id
):
65 webpage_url
= self
._MOBILE
_TEMPLATE
% mtvn_id
66 req
= sanitized_Request(webpage_url
)
67 # Otherwise we get a webpage that would execute some javascript
68 req
.add_header('User-Agent', 'curl/7')
69 webpage
= self
._download
_webpage
(req
, mtvn_id
,
70 'Downloading mobile page')
71 metrics_url
= unescapeHTML(self
._search
_regex
(r
'<a href="(http://metrics.+?)"', webpage
, 'url'))
72 req
= HEADRequest(metrics_url
)
73 response
= self
._request
_webpage
(req
, mtvn_id
, 'Resolving url')
74 url
= response
.geturl()
75 # Transform the url to get the best quality:
76 url
= re
.sub(r
'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url
, 1)
77 return [{'url': url, 'ext': 'mp4'}
]
79 def _extract_video_formats(self
, mdoc
, mtvn_id
):
80 if re
.match(r
'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc
.find('.//src').text
) is not None:
81 if mtvn_id
is not None and self
._MOBILE
_TEMPLATE
is not None:
82 self
.to_screen('The normal version is not available from your '
83 'country, trying with the mobile version')
84 return self
._extract
_mobile
_video
_formats
(mtvn_id
)
85 raise ExtractorError('This video is not available from your country.',
89 for rendition
in mdoc
.findall('.//rendition'):
91 _
, _
, ext
= rendition
.attrib
['type'].partition('/')
92 rtmp_video_url
= rendition
.find('./src').text
93 if rtmp_video_url
.endswith('siteunavail.png'):
95 new_urls
= self
._transform
_rtmp
_url
(rtmp_video_url
)
97 'ext': 'flv' if new_url
.startswith('rtmp') else ext
,
99 'format_id': '-'.join(filter(None, [kind
, rendition
.get('bitrate')])),
100 'width': int(rendition
.get('width')),
101 'height': int(rendition
.get('height')),
102 } for kind
, new_url
in new_urls
.items()])
103 except (KeyError, TypeError):
104 raise ExtractorError('Invalid rendition field.')
105 self
._sort
_formats
(formats
)
108 def _extract_subtitles(self
, mdoc
, mtvn_id
):
110 for transcript
in mdoc
.findall('.//transcript'):
111 if transcript
.get('kind') != 'captions':
113 lang
= transcript
.get('srclang')
115 'url': compat_str(typographic
.get('src')),
116 'ext': typographic
.get('format')
117 } for typographic
in transcript
.findall('./typographic')]
120 def _get_video_info(self
, itemdoc
):
121 uri
= itemdoc
.find('guid').text
122 video_id
= self
._id
_from
_uri
(uri
)
123 self
.report_extraction(video_id
)
124 content_el
= itemdoc
.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content')))
125 mediagen_url
= self
._remove
_template
_parameter
(content_el
.attrib
['url'])
126 if 'acceptMethods' not in mediagen_url
:
127 mediagen_url
+= '&' if '?' in mediagen_url
else '?'
128 mediagen_url
+= 'acceptMethods=fms'
130 mediagen_doc
= self
._download
_xml
(mediagen_url
, video_id
,
131 'Downloading video urls')
133 item
= mediagen_doc
.find('./video/item')
134 if item
is not None and item
.get('type') == 'text':
135 message
= '%s returned error: ' % self
.IE_NAME
136 if item
.get('code') is not None:
137 message
+= '%s - ' % item
.get('code')
139 raise ExtractorError(message
, expected
=True)
141 description
= strip_or_none(xpath_text(itemdoc
, 'description'))
143 timestamp
= timeconvert(xpath_text(itemdoc
, 'pubDate'))
147 title_el
= find_xpath_attr(
148 itemdoc
, './/{http://search.yahoo.com/mrss/}category',
149 'scheme', 'urn:mtvn:video_title')
151 title_el
= itemdoc
.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title'))
153 title_el
= itemdoc
.find(compat_xpath('.//title'))
154 if title_el
.text
is None:
157 title
= title_el
.text
159 raise ExtractorError('Could not find video title')
160 title
= title
.strip()
162 # This a short id that's used in the webpage urls
164 mtvn_id_node
= find_xpath_attr(itemdoc
, './/{http://search.yahoo.com/mrss/}category',
165 'scheme', 'urn:mtvn:id')
166 if mtvn_id_node
is not None:
167 mtvn_id
= mtvn_id_node
.text
171 'formats': self
._extract
_video
_formats
(mediagen_doc
, mtvn_id
),
172 'subtitles': self
._extract
_subtitles
(mediagen_doc
, mtvn_id
),
174 'thumbnail': self
._get
_thumbnail
_url
(uri
, itemdoc
),
175 'description': description
,
176 'duration': float_or_none(content_el
.attrib
.get('duration')),
177 'timestamp': timestamp
,
180 def _get_feed_query(self
, uri
):
183 data
['lang'] = self
._LANG
186 def _get_videos_info(self
, uri
):
187 video_id
= self
._id
_from
_uri
(uri
)
188 feed_url
= self
._get
_feed
_url
(uri
)
189 info_url
= update_url_query(feed_url
, self
._get
_feed
_query
(uri
))
190 return self
._get
_videos
_info
_from
_url
(info_url
, video_id
)
192 def _get_videos_info_from_url(self
, url
, video_id
):
193 idoc
= self
._download
_xml
(
195 'Downloading info', transform_source
=fix_xml_ampersands
)
197 title
= xpath_text(idoc
, './channel/title')
198 description
= xpath_text(idoc
, './channel/description')
200 return self
.playlist_result(
201 [self
._get
_video
_info
(item
) for item
in idoc
.findall('.//item')],
202 playlist_title
=title
, playlist_description
=description
)
204 def _extract_mgid(self
, webpage
):
206 # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
207 # or http://media.mtvnservices.com/{mgid}
208 og_url
= self
._og
_search
_video
_url
(webpage
)
209 mgid
= url_basename(og_url
)
210 if mgid
.endswith('.swf'):
212 except RegexNotFoundError
:
215 if mgid
is None or ':' not in mgid
:
216 mgid
= self
._search
_regex
(
217 [r
'data-mgid="(.*?)"', r
'swfobject.embedSWF\(".*?(mgid:.*?)"'],
218 webpage
, 'mgid', default
=None)
221 sm4_embed
= self
._html
_search
_meta
(
222 'sm4:video:embed', webpage
, 'sm4 embed', default
='')
223 mgid
= self
._search
_regex
(
224 r
'embed/(mgid:.+?)["\'&?
/]', sm4_embed, 'mgid
')
227 def _real_extract(self, url):
228 title = url_basename(url)
229 webpage = self._download_webpage(url, title)
230 mgid = self._extract_mgid(webpage)
231 videos_info = self._get_videos_info(mgid)
235 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
236 IE_NAME = 'mtvservices
:embedded
'
237 _VALID_URL = r'https?
://media\
.mtvnservices\
.com
/embed
/(?P
<mgid
>.+?
)(\?|
/|$
)'
240 # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
241 'url
': 'http
://media
.mtvnservices
.com
/embed
/mgid
:uma
:video
:mtv
.com
:1043906/cp~vid
%3D1043906
%26uri
%3Dmgid
%3Auma
%3Avideo
%3Amtv
.com
%3A1043906
',
242 'md5
': 'cb349b21a7897164cede95bd7bf3fbb9
',
246 'title
': 'Peter Dinklage Sums Up
\'Game Of Thrones
\' In
45 Seconds
',
247 'description
': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage
as he tries summarizing
"Game of Thrones" in under a minute
.',
248 'timestamp
': 1400126400,
249 'upload_date
': '20140515',
254 def _extract_url(webpage):
256 r'<iframe
[^
>]+?src
=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage)
258 return mobj.group('url')
260 def _get_feed_url(self, uri):
261 video_id = self._id_from_uri(uri)
262 config = self._download_json(
263 'http://media.mtvnservices.com/pmt/e1/access/index.html?uri=%s&configtype=edge' % uri, video_id)
264 return self._remove_template_parameter(config['feedWithQueryParams'])
266 def _real_extract(self, url):
267 mobj = re.match(self._VALID_URL, url)
268 mgid = mobj.group('mgid')
269 return self._get_videos_info(mgid)
272 class MTVIE(MTVServicesInfoExtractor):
273 _VALID_URL = r'''(?x)^https?://
274 (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
275 m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
277 _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
281 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
282 'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
286 'title': 'Taylor Swift - "Ours (VH1 Storytellers
)"',
287 'description': 'Album: Taylor Swift performs "Ours
" for VH1 Storytellers at Harvey Mudd College.',
288 'timestamp': 1352610000,
289 'upload_date': '20121111',
294 def _get_thumbnail_url(self, uri, itemdoc):
295 return 'http://mtv.mtvnimages.com/uri/' + uri
297 def _real_extract(self, url):
298 mobj = re.match(self._VALID_URL, url)
299 video_id = mobj.group('videoid')
300 uri = mobj.groupdict().get('mgid')
302 webpage = self._download_webpage(url, video_id)
304 # Some videos come from Vevo.com
306 r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?
)";', webpage)
308 vevo_id = m_vevo.group(1)
309 self.to_screen('Vevo video detected: %s' % vevo_id)
310 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
312 uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
313 return self._get_videos_info(uri)
316 class MTVDEIE(MTVServicesInfoExtractor):
318 _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$'
320 'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
322 'id': 'music_video-a50bc5f0b3aa4b3190aa',
324 'title': 'MusicVideo_cro-traum',
325 'description': 'Cro - Traum',
329 'skip_download': True,
331 'skip': 'Blocked at Travis CI',
333 # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
334 'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen',
336 'id': 'local_playlist-f5ae778b9832cc837189',
338 'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
342 'skip_download': True,
344 'skip': 'Blocked at Travis CI',
346 'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
348 'id': 'local_playlist-4e760566473c4c8c5344',
350 'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
351 'description': 'MTV Movies Supercut',
355 'skip_download': True,
357 'skip': 'Das Video kann zur Zeit nicht abgespielt werden.',
360 def _real_extract(self, url):
361 video_id = self._match_id(url)
363 webpage = self._download_webpage(url, video_id)
365 playlist = self._parse_json(
367 r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'),
371 return item['mrss'] + item.get('mrssvars', '')
373 # news pages contain single video in playlist with different id
374 if len(playlist) == 1:
375 return self._get_videos_info_from_url(_mrss_url(playlist[0]), video_id)
377 for item in playlist:
378 item_id = item.get('id')
379 if item_id and compat_str(item_id) == video_id:
380 return self._get_videos_info_from_url(_mrss_url(item), video_id)