1 from __future__
import unicode_literals
5 from .common
import InfoExtractor
28 def _media_xml_tag(tag
):
29 return '{http://search.yahoo.com/mrss/}%s' % tag
32 class MTVServicesInfoExtractor(InfoExtractor
):
33 _MOBILE_TEMPLATE
= None
37 def _id_from_uri(uri
):
38 return uri
.split(':')[-1]
41 def _remove_template_parameter(url
):
42 # Remove the templates, like &device={device}
43 return re
.sub(r
'&[^=]*?={.*?}(?=(&|$))', '', url
)
45 # This was originally implemented for ComedyCentral, but it also works here
47 def _transform_rtmp_url(cls
, rtmp_video_url
):
48 m
= re
.match(r
'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url
)
50 return {'rtmp': rtmp_video_url}
51 base
= 'http://viacommtvstrmfs.fplive.net/'
52 return {'http': base + m.group('finalid')}
54 def _get_feed_url(self
, uri
):
57 def _get_thumbnail_url(self
, uri
, itemdoc
):
58 search_path
= '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
59 thumb_node
= itemdoc
.find(search_path
)
60 if thumb_node
is None:
63 return thumb_node
.attrib
['url']
65 def _extract_mobile_video_formats(self
, mtvn_id
):
66 webpage_url
= self
._MOBILE
_TEMPLATE
% mtvn_id
67 req
= sanitized_Request(webpage_url
)
68 # Otherwise we get a webpage that would execute some javascript
69 req
.add_header('User-Agent', 'curl/7')
70 webpage
= self
._download
_webpage
(req
, mtvn_id
,
71 'Downloading mobile page')
72 metrics_url
= unescapeHTML(self
._search
_regex
(r
'<a href="(http://metrics.+?)"', webpage
, 'url'))
73 req
= HEADRequest(metrics_url
)
74 response
= self
._request
_webpage
(req
, mtvn_id
, 'Resolving url')
75 url
= response
.geturl()
76 # Transform the url to get the best quality:
77 url
= re
.sub(r
'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url
, 1)
78 return [{'url': url, 'ext': 'mp4'}
]
80 def _extract_video_formats(self
, mdoc
, mtvn_id
, video_id
):
81 if re
.match(r
'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc
.find('.//src').text
) is not None:
82 if mtvn_id
is not None and self
._MOBILE
_TEMPLATE
is not None:
83 self
.to_screen('The normal version is not available from your '
84 'country, trying with the mobile version')
85 return self
._extract
_mobile
_video
_formats
(mtvn_id
)
86 raise ExtractorError('This video is not available from your country.',
90 for rendition
in mdoc
.findall('.//rendition'):
91 if rendition
.get('method') == 'hls':
92 hls_url
= rendition
.find('./src').text
93 formats
.extend(self
._extract
_m
3u8_formats
(hls_url
, video_id
, ext
='mp4'))
97 _
, _
, ext
= rendition
.attrib
['type'].partition('/')
98 rtmp_video_url
= rendition
.find('./src').text
99 if rtmp_video_url
.endswith('siteunavail.png'):
101 new_urls
= self
._transform
_rtmp
_url
(rtmp_video_url
)
103 'ext': 'flv' if new_url
.startswith('rtmp') else ext
,
105 'format_id': '-'.join(filter(None, [kind
, rendition
.get('bitrate')])),
106 'width': int(rendition
.get('width')),
107 'height': int(rendition
.get('height')),
108 } for kind
, new_url
in new_urls
.items()])
109 except (KeyError, TypeError):
110 raise ExtractorError('Invalid rendition field.')
111 self
._sort
_formats
(formats
)
114 def _extract_subtitles(self
, mdoc
, mtvn_id
):
116 for transcript
in mdoc
.findall('.//transcript'):
117 if transcript
.get('kind') != 'captions':
119 lang
= transcript
.get('srclang')
121 'url': compat_str(typographic
.get('src')),
122 'ext': typographic
.get('format')
123 } for typographic
in transcript
.findall('./typographic')]
126 def _get_video_info(self
, itemdoc
, use_hls
=True):
127 uri
= itemdoc
.find('guid').text
128 video_id
= self
._id
_from
_uri
(uri
)
129 self
.report_extraction(video_id
)
130 content_el
= itemdoc
.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content')))
131 mediagen_url
= self
._remove
_template
_parameter
(content_el
.attrib
['url'])
132 mediagen_url
= mediagen_url
.replace('device={device}', '')
133 if 'acceptMethods' not in mediagen_url
:
134 mediagen_url
+= '&' if '?' in mediagen_url
else '?'
135 mediagen_url
+= 'acceptMethods='
136 mediagen_url
+= 'hls' if use_hls
else 'fms'
138 mediagen_doc
= self
._download
_xml
(mediagen_url
, video_id
,
139 'Downloading video urls')
141 item
= mediagen_doc
.find('./video/item')
142 if item
is not None and item
.get('type') == 'text':
143 message
= '%s returned error: ' % self
.IE_NAME
144 if item
.get('code') is not None:
145 message
+= '%s - ' % item
.get('code')
147 raise ExtractorError(message
, expected
=True)
149 description
= strip_or_none(xpath_text(itemdoc
, 'description'))
151 timestamp
= timeconvert(xpath_text(itemdoc
, 'pubDate'))
155 title_el
= find_xpath_attr(
156 itemdoc
, './/{http://search.yahoo.com/mrss/}category',
157 'scheme', 'urn:mtvn:video_title')
159 title_el
= itemdoc
.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title'))
161 title_el
= itemdoc
.find(compat_xpath('.//title'))
162 if title_el
.text
is None:
165 title
= title_el
.text
167 raise ExtractorError('Could not find video title')
168 title
= title
.strip()
170 # This a short id that's used in the webpage urls
172 mtvn_id_node
= find_xpath_attr(itemdoc
, './/{http://search.yahoo.com/mrss/}category',
173 'scheme', 'urn:mtvn:id')
174 if mtvn_id_node
is not None:
175 mtvn_id
= mtvn_id_node
.text
177 formats
= self
._extract
_video
_formats
(mediagen_doc
, mtvn_id
, video_id
)
182 'subtitles': self
._extract
_subtitles
(mediagen_doc
, mtvn_id
),
184 'thumbnail': self
._get
_thumbnail
_url
(uri
, itemdoc
),
185 'description': description
,
186 'duration': float_or_none(content_el
.attrib
.get('duration')),
187 'timestamp': timestamp
,
190 def _get_feed_query(self
, uri
):
193 data
['lang'] = self
._LANG
196 def _get_videos_info(self
, uri
, use_hls
=True):
197 video_id
= self
._id
_from
_uri
(uri
)
198 feed_url
= self
._get
_feed
_url
(uri
)
199 info_url
= update_url_query(feed_url
, self
._get
_feed
_query
(uri
))
200 return self
._get
_videos
_info
_from
_url
(info_url
, video_id
, use_hls
)
202 def _get_videos_info_from_url(self
, url
, video_id
, use_hls
=True):
203 idoc
= self
._download
_xml
(
205 'Downloading info', transform_source
=fix_xml_ampersands
)
207 title
= xpath_text(idoc
, './channel/title')
208 description
= xpath_text(idoc
, './channel/description')
210 return self
.playlist_result(
211 [self
._get
_video
_info
(item
, use_hls
) for item
in idoc
.findall('.//item')],
212 playlist_title
=title
, playlist_description
=description
)
214 def _extract_mgid(self
, webpage
, default
=NO_DEFAULT
):
216 # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
217 # or http://media.mtvnservices.com/{mgid}
218 og_url
= self
._og
_search
_video
_url
(webpage
)
219 mgid
= url_basename(og_url
)
220 if mgid
.endswith('.swf'):
222 except RegexNotFoundError
:
225 if mgid
is None or ':' not in mgid
:
226 mgid
= self
._search
_regex
(
227 [r
'data-mgid="(.*?)"', r
'swfobject.embedSWF\(".*?(mgid:.*?)"'],
228 webpage
, 'mgid', default
=None)
231 sm4_embed
= self
._html
_search
_meta
(
232 'sm4:video:embed', webpage
, 'sm4 embed', default
='')
233 mgid
= self
._search
_regex
(
234 r
'embed/(mgid:.+?)["\'&?
/]', sm4_embed, 'mgid
', default=default)
237 def _real_extract(self, url):
238 title = url_basename(url)
239 webpage = self._download_webpage(url, title)
240 mgid = self._extract_mgid(webpage)
241 videos_info = self._get_videos_info(mgid)
245 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
246 IE_NAME = 'mtvservices
:embedded
'
247 _VALID_URL = r'https?
://media\
.mtvnservices\
.com
/embed
/(?P
<mgid
>.+?
)(\?|
/|$
)'
250 # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
251 'url
': 'http
://media
.mtvnservices
.com
/embed
/mgid
:uma
:video
:mtv
.com
:1043906/cp~vid
%3D1043906
%26uri
%3Dmgid
%3Auma
%3Avideo
%3Amtv
.com
%3A1043906
',
252 'md5
': 'cb349b21a7897164cede95bd7bf3fbb9
',
256 'title
': 'Peter Dinklage Sums Up
\'Game Of Thrones
\' In
45 Seconds
',
257 'description
': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage
as he tries summarizing
"Game of Thrones" in under a minute
.',
258 'timestamp
': 1400126400,
259 'upload_date
': '20140515',
264 def _extract_url(webpage):
266 r'<iframe
[^
>]+?src
=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage)
268 return mobj.group('url')
270 def _get_feed_url(self, uri):
271 video_id = self._id_from_uri(uri)
272 config = self._download_json(
273 'http://media.mtvnservices.com/pmt/e1/access/index.html?uri=%s&configtype=edge' % uri, video_id)
274 return self._remove_template_parameter(config['feedWithQueryParams'])
276 def _real_extract(self, url):
277 mobj = re.match(self._VALID_URL, url)
278 mgid = mobj.group('mgid')
279 return self._get_videos_info(mgid)
282 class MTVIE(MTVServicesInfoExtractor):
284 _VALID_URL = r'https?://(?:www\.)?mtv\.com/(?:video-clips|full-episodes)/(?P<id>[^/?#.]+)'
285 _FEED_URL = 'http://www.mtv.com/feeds/mrss/'
288 'url': 'http://www.mtv.com/video-clips/vl8qof/unlocking-the-truth-trailer',
289 'md5': '1edbcdf1e7628e414a8c5dcebca3d32b',
291 'id': '5e14040d-18a4-47c4-a582-43ff602de88e',
293 'title': 'Unlocking The Truth|July 18, 2016|1|101|Trailer',
294 'description': '"Unlocking the Truth
" premieres August 17th at 11/10c.',
295 'timestamp': 1468846800,
296 'upload_date': '20160718',
299 'url': 'http://www.mtv.com/full-episodes/94tujl/unlocking-the-truth-gates-of-hell-season-1-ep-101',
300 'only_matching': True,
304 class MTVVideoIE(MTVServicesInfoExtractor):
305 IE_NAME = 'mtv:video'
306 _VALID_URL = r'''(?x)^https?://
307 (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
308 m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
310 _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
314 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
315 'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
319 'title': 'Taylor Swift - "Ours (VH1 Storytellers
)"',
320 'description': 'Album: Taylor Swift performs "Ours
" for VH1 Storytellers at Harvey Mudd College.',
321 'timestamp': 1352610000,
322 'upload_date': '20121111',
327 def _get_thumbnail_url(self, uri, itemdoc):
328 return 'http://mtv.mtvnimages.com/uri/' + uri
330 def _real_extract(self, url):
331 mobj = re.match(self._VALID_URL, url)
332 video_id = mobj.group('videoid')
333 uri = mobj.groupdict().get('mgid')
335 webpage = self._download_webpage(url, video_id)
337 # Some videos come from Vevo.com
339 r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?
)";', webpage)
341 vevo_id = m_vevo.group(1)
342 self.to_screen('Vevo video detected: %s' % vevo_id)
343 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
345 uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
346 return self._get_videos_info(uri)
349 class MTVDEIE(MTVServicesInfoExtractor):
351 _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$'
353 'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
355 'id': 'music_video-a50bc5f0b3aa4b3190aa',
357 'title': 'MusicVideo_cro-traum',
358 'description': 'Cro - Traum',
362 'skip_download': True,
364 'skip': 'Blocked at Travis CI',
366 # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
367 'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen',
369 'id': 'local_playlist-f5ae778b9832cc837189',
371 'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
375 'skip_download': True,
377 'skip': 'Blocked at Travis CI',
379 'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
381 'id': 'local_playlist-4e760566473c4c8c5344',
383 'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
384 'description': 'MTV Movies Supercut',
388 'skip_download': True,
390 'skip': 'Das Video kann zur Zeit nicht abgespielt werden.',
393 def _real_extract(self, url):
394 video_id = self._match_id(url)
396 webpage = self._download_webpage(url, video_id)
398 playlist = self._parse_json(
400 r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'),
404 return item['mrss'] + item.get('mrssvars', '')
406 # news pages contain single video in playlist with different id
407 if len(playlist) == 1:
408 return self._get_videos_info_from_url(_mrss_url(playlist[0]), video_id)
410 for item in playlist:
411 item_id = item.get('id')
412 if item_id and compat_str(item_id) == video_id:
413 return self._get_videos_info_from_url(_mrss_url(item), video_id)