2 from __future__
import unicode_literals
6 from .common
import InfoExtractor
31 def _media_xml_tag(tag
):
32 return '{http://search.yahoo.com/mrss/}%s' % tag
35 class MTVServicesInfoExtractor(InfoExtractor
):
36 _MOBILE_TEMPLATE
= None
40 def _id_from_uri(uri
):
41 return uri
.split(':')[-1]
44 def _remove_template_parameter(url
):
45 # Remove the templates, like &device={device}
46 return re
.sub(r
'&[^=]*?={.*?}(?=(&|$))', '', url
)
48 def _get_feed_url(self
, uri
, url
=None):
51 def _get_thumbnail_url(self
, uri
, itemdoc
):
52 search_path
= '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
53 thumb_node
= itemdoc
.find(search_path
)
54 if thumb_node
is None:
56 return thumb_node
.get('url') or thumb_node
.text
or None
58 def _extract_mobile_video_formats(self
, mtvn_id
):
59 webpage_url
= self
._MOBILE
_TEMPLATE
% mtvn_id
60 req
= sanitized_Request(webpage_url
)
61 # Otherwise we get a webpage that would execute some javascript
62 req
.add_header('User-Agent', 'curl/7')
63 webpage
= self
._download
_webpage
(req
, mtvn_id
,
64 'Downloading mobile page')
65 metrics_url
= unescapeHTML(self
._search
_regex
(r
'<a href="(http://metrics.+?)"', webpage
, 'url'))
66 req
= HEADRequest(metrics_url
)
67 response
= self
._request
_webpage
(req
, mtvn_id
, 'Resolving url')
68 url
= response
.geturl()
69 # Transform the url to get the best quality:
70 url
= re
.sub(r
'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url
, 1)
71 return [{'url': url, 'ext': 'mp4'}
]
73 def _extract_video_formats(self
, mdoc
, mtvn_id
, video_id
):
74 if re
.match(r
'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc
.find('.//src').text
) is not None:
75 if mtvn_id
is not None and self
._MOBILE
_TEMPLATE
is not None:
76 self
.to_screen('The normal version is not available from your '
77 'country, trying with the mobile version')
78 return self
._extract
_mobile
_video
_formats
(mtvn_id
)
79 raise ExtractorError('This video is not available from your country.',
83 for rendition
in mdoc
.findall('.//rendition'):
84 if rendition
.get('method') == 'hls':
85 hls_url
= rendition
.find('./src').text
86 formats
.extend(self
._extract
_m
3u8_formats
(
87 hls_url
, video_id
, ext
='mp4', entry_protocol
='m3u8_native',
88 m3u8_id
='hls', fatal
=False))
92 _
, _
, ext
= rendition
.attrib
['type'].partition('/')
93 rtmp_video_url
= rendition
.find('./src').text
94 if 'error_not_available.swf' in rtmp_video_url
:
96 '%s said: video is not available' % self
.IE_NAME
,
98 if rtmp_video_url
.endswith('siteunavail.png'):
101 'ext': 'flv' if rtmp_video_url
.startswith('rtmp') else ext
,
102 'url': rtmp_video_url
,
103 'format_id': join_nonempty(
104 'rtmp' if rtmp_video_url
.startswith('rtmp') else None,
105 rendition
.get('bitrate')),
106 'width': int(rendition
.get('width')),
107 'height': int(rendition
.get('height')),
109 except (KeyError, TypeError):
110 raise ExtractorError('Invalid rendition field.')
112 self
._sort
_formats
(formats
)
115 def _extract_subtitles(self
, mdoc
, mtvn_id
):
117 for transcript
in mdoc
.findall('.//transcript'):
118 if transcript
.get('kind') != 'captions':
120 lang
= transcript
.get('srclang')
121 for typographic
in transcript
.findall('./typographic'):
122 sub_src
= typographic
.get('src')
125 ext
= typographic
.get('format')
128 subtitles
.setdefault(lang
, []).append({
129 'url': compat_str(sub_src
),
134 def _get_video_info(self
, itemdoc
, use_hls
=True):
135 uri
= itemdoc
.find('guid').text
136 video_id
= self
._id
_from
_uri
(uri
)
137 self
.report_extraction(video_id
)
138 content_el
= itemdoc
.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content')))
139 mediagen_url
= self
._remove
_template
_parameter
(content_el
.attrib
['url'])
140 mediagen_url
= mediagen_url
.replace('device={device}', '')
141 if 'acceptMethods' not in mediagen_url
:
142 mediagen_url
+= '&' if '?' in mediagen_url
else '?'
143 mediagen_url
+= 'acceptMethods='
144 mediagen_url
+= 'hls' if use_hls
else 'fms'
146 mediagen_doc
= self
._download
_xml
(
147 mediagen_url
, video_id
, 'Downloading video urls', fatal
=False)
149 if mediagen_doc
is False:
152 item
= mediagen_doc
.find('./video/item')
153 if item
is not None and item
.get('type') == 'text':
154 message
= '%s returned error: ' % self
.IE_NAME
155 if item
.get('code') is not None:
156 message
+= '%s - ' % item
.get('code')
158 raise ExtractorError(message
, expected
=True)
160 description
= strip_or_none(xpath_text(itemdoc
, 'description'))
162 timestamp
= timeconvert(xpath_text(itemdoc
, 'pubDate'))
166 title_el
= find_xpath_attr(
167 itemdoc
, './/{http://search.yahoo.com/mrss/}category',
168 'scheme', 'urn:mtvn:video_title')
170 title_el
= itemdoc
.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title'))
172 title_el
= itemdoc
.find(compat_xpath('.//title'))
173 if title_el
.text
is None:
176 title
= title_el
.text
178 raise ExtractorError('Could not find video title')
179 title
= title
.strip()
181 series
= find_xpath_attr(
182 itemdoc
, './/{http://search.yahoo.com/mrss/}category',
183 'scheme', 'urn:mtvn:franchise')
184 season
= find_xpath_attr(
185 itemdoc
, './/{http://search.yahoo.com/mrss/}category',
186 'scheme', 'urn:mtvn:seasonN')
187 episode
= find_xpath_attr(
188 itemdoc
, './/{http://search.yahoo.com/mrss/}category',
189 'scheme', 'urn:mtvn:episodeN')
190 series
= series
.text
if series
is not None else None
191 season
= season
.text
if season
is not None else None
192 episode
= episode
.text
if episode
is not None else None
193 if season
and episode
:
194 # episode number includes season, so remove it
195 episode
= re
.sub(r
'^%s' % season
, '', episode
)
197 # This a short id that's used in the webpage urls
199 mtvn_id_node
= find_xpath_attr(itemdoc
, './/{http://search.yahoo.com/mrss/}category',
200 'scheme', 'urn:mtvn:id')
201 if mtvn_id_node
is not None:
202 mtvn_id
= mtvn_id_node
.text
204 formats
= self
._extract
_video
_formats
(mediagen_doc
, mtvn_id
, video_id
)
206 # Some parts of complete video may be missing (e.g. missing Act 3 in
207 # http://www.southpark.de/alle-episoden/s14e01-sexual-healing)
211 self
._sort
_formats
(formats
)
216 'subtitles': self
._extract
_subtitles
(mediagen_doc
, mtvn_id
),
218 'thumbnail': self
._get
_thumbnail
_url
(uri
, itemdoc
),
219 'description': description
,
220 'duration': float_or_none(content_el
.attrib
.get('duration')),
221 'timestamp': timestamp
,
223 'season_number': int_or_none(season
),
224 'episode_number': int_or_none(episode
),
227 def _get_feed_query(self
, uri
):
230 data
['lang'] = self
._LANG
233 def _get_videos_info(self
, uri
, use_hls
=True, url
=None):
234 video_id
= self
._id
_from
_uri
(uri
)
235 feed_url
= self
._get
_feed
_url
(uri
, url
)
236 info_url
= update_url_query(feed_url
, self
._get
_feed
_query
(uri
))
237 return self
._get
_videos
_info
_from
_url
(info_url
, video_id
, use_hls
)
239 def _get_videos_info_from_url(self
, url
, video_id
, use_hls
=True):
240 idoc
= self
._download
_xml
(
242 'Downloading info', transform_source
=fix_xml_ampersands
)
244 title
= xpath_text(idoc
, './channel/title')
245 description
= xpath_text(idoc
, './channel/description')
248 for item
in idoc
.findall('.//item'):
249 info
= self
._get
_video
_info
(item
, use_hls
)
253 # TODO: should be multi-video
254 return self
.playlist_result(
255 entries
, playlist_title
=title
, playlist_description
=description
)
257 def _extract_triforce_mgid(self
, webpage
, data_zone
=None, video_id
=None):
258 triforce_feed
= self
._parse
_json
(self
._search
_regex
(
259 r
'triforceManifestFeed\s*=\s*({.+?})\s*;\s*\n', webpage
,
260 'triforce feed', default
='{}'), video_id
, fatal
=False)
262 data_zone
= self
._search
_regex
(
263 r
'data-zone=(["\'])(?P
<zone
>.+?_lc_promo
.*?
)\
1', webpage,
264 'data zone
', default=data_zone, group='zone
')
267 triforce_feed, lambda x: x['manifest
']['zones
'][data_zone]['feed
'],
272 feed = self._download_json(feed_url, video_id, fatal=False)
276 return try_get(feed, lambda x: x['result
']['data
']['id'], compat_str)
279 def _extract_child_with_type(parent, t):
280 for c in parent['children
']:
281 if c.get('type') == t:
284 def _extract_mgid(self, webpage):
286 # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
287 # or http://media.mtvnservices.com/{mgid}
288 og_url = self._og_search_video_url(webpage)
289 mgid = url_basename(og_url)
290 if mgid.endswith('.swf
'):
292 except RegexNotFoundError:
295 if mgid is None or ':' not in mgid:
296 mgid = self._search_regex(
297 [r'data
-mgid
="(.*?)"', r'swfobject\
.embedSWF\
(".*?(mgid:.*?)"'],
298 webpage, 'mgid
', default=None)
301 sm4_embed = self._html_search_meta(
302 'sm4
:video
:embed
', webpage, 'sm4 embed
', default='')
303 mgid = self._search_regex(
304 r'embed
/(mgid
:.+?
)["\'&?/]', sm4_embed, 'mgid', default=None)
307 mgid = self._extract_triforce_mgid(webpage)
310 mgid = self._search_regex(
311 r'"videoConfig
":{"videoId
":"(mgid
:.*?
)"', webpage, 'mgid', default=None)
314 mgid = self._search_regex(
315 r'"media
":{"video
":{"config
":{"uri
":"(mgid
:.*?
)"', webpage, 'mgid', default=None)
318 data = self._parse_json(self._search_regex(
319 r'__DATA__\s*=\s*({.+?});', webpage, 'data'), None)
320 main_container = self._extract_child_with_type(data, 'MainContainer')
321 ab_testing = self._extract_child_with_type(main_container, 'ABTesting')
322 video_player = self._extract_child_with_type(ab_testing or main_container, 'VideoPlayer')
323 mgid = video_player['props']['media']['video']['config']['uri']
327 def _real_extract(self, url):
328 title = url_basename(url)
329 webpage = self._download_webpage(url, title)
330 mgid = self._extract_mgid(webpage)
331 videos_info = self._get_videos_info(mgid, url=url)
335 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
336 IE_NAME = 'mtvservices:embedded'
337 _VALID_URL = r'https?://media\.mtvnservices\.com/embed/(?P<mgid>.+?)(\?|/|$)'
340 # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
341 'url': 'http://media.mtvnservices.com/embed/mgid:uma:video:mtv.com:1043906/cp~vid%3D1043906%26uri%3Dmgid%3Auma%3Avideo%3Amtv.com%3A1043906',
342 'md5': 'cb349b21a7897164cede95bd7bf3fbb9',
346 'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds',
347 'description': '"Sexy sexy sexy
, stabby stabby stabby
, beautiful language
," says Peter Dinklage as he tries summarizing "Game of Thrones
" in under a minute.',
348 'timestamp': 1400126400,
349 'upload_date': '20140515',
354 def _extract_url(webpage):
356 r'<iframe[^>]+?src=(["\'])(?P
<url
>(?
:https?
:)?
//media\
.mtvnservices\
.com
/embed
/.+?
)\
1', webpage)
358 return mobj.group('url
')
360 def _get_feed_url(self, uri, url=None):
361 video_id = self._id_from_uri(uri)
362 config = self._download_json(
363 'http
://media
.mtvnservices
.com
/pmt
/e1
/access
/index
.html?uri
=%s&configtype
=edge
' % uri, video_id)
364 return self._remove_template_parameter(config['feedWithQueryParams
'])
366 def _real_extract(self, url):
367 mobj = self._match_valid_url(url)
368 mgid = mobj.group('mgid
')
369 return self._get_videos_info(mgid)
372 class MTVIE(MTVServicesInfoExtractor):
374 _VALID_URL = r'https?
://(?
:www\
.)?mtv\
.com
/(?
:video
-clips|
(?
:full
-)?episodes
)/(?P
<id>[^
/?
#.]+)'
375 _FEED_URL
= 'http://www.mtv.com/feeds/mrss/'
378 'url': 'http://www.mtv.com/video-clips/vl8qof/unlocking-the-truth-trailer',
379 'md5': '1edbcdf1e7628e414a8c5dcebca3d32b',
381 'id': '5e14040d-18a4-47c4-a582-43ff602de88e',
383 'title': 'Unlocking The Truth|July 18, 2016|1|101|Trailer',
384 'description': '"Unlocking the Truth" premieres August 17th at 11/10c.',
385 'timestamp': 1468846800,
386 'upload_date': '20160718',
389 'url': 'http://www.mtv.com/full-episodes/94tujl/unlocking-the-truth-gates-of-hell-season-1-ep-101',
390 'only_matching': True,
392 'url': 'http://www.mtv.com/episodes/g8xu7q/teen-mom-2-breaking-the-wall-season-7-ep-713',
393 'only_matching': True,
397 class MTVJapanIE(MTVServicesInfoExtractor
):
399 _VALID_URL
= r
'https?://(?:www\.)?mtvjapan\.com/videos/(?P<id>[0-9a-z]+)'
402 'url': 'http://www.mtvjapan.com/videos/prayht/fresh-info-cadillac-escalade',
404 'id': 'bc01da03-6fe5-4284-8880-f291f4e368f5',
406 'title': '【Fresh Info】Cadillac ESCALADE Sport Edition',
409 'skip_download': True,
412 _GEO_COUNTRIES
= ['JP']
413 _FEED_URL
= 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
415 def _get_feed_query(self
, uri
):
417 'arcEp': 'mtvjapan.com',
422 class MTVVideoIE(MTVServicesInfoExtractor
):
423 IE_NAME
= 'mtv:video'
424 _VALID_URL
= r
'''(?x)^https?://
425 (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
426 m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
428 _FEED_URL
= 'http://www.mtv.com/player/embed/AS3/rss/'
432 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
433 'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
437 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
438 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
439 'timestamp': 1352610000,
440 'upload_date': '20121111',
445 def _get_thumbnail_url(self
, uri
, itemdoc
):
446 return 'http://mtv.mtvnimages.com/uri/' + uri
448 def _real_extract(self
, url
):
449 mobj
= self
._match
_valid
_url
(url
)
450 video_id
= mobj
.group('videoid')
451 uri
= mobj
.groupdict().get('mgid')
453 webpage
= self
._download
_webpage
(url
, video_id
)
455 # Some videos come from Vevo.com
457 r
'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage
)
459 vevo_id
= m_vevo
.group(1)
460 self
.to_screen('Vevo video detected: %s' % vevo_id
)
461 return self
.url_result('vevo:%s' % vevo_id
, ie
='Vevo')
463 uri
= self
._html
_search
_regex
(r
'/uri/(.*?)\?', webpage
, 'uri')
464 return self
._get
_videos
_info
(uri
)
467 class MTVDEIE(MTVServicesInfoExtractor
):
469 _VALID_URL
= r
'https?://(?:www\.)?mtv\.de/(?:musik/videoclips|folgen|news)/(?P<id>[0-9a-z]+)'
471 'url': 'http://www.mtv.de/musik/videoclips/2gpnv7/Traum',
473 'id': 'd5d472bc-f5b7-11e5-bffd-a4badb20dab5',
476 'description': 'Traum',
480 'skip_download': True,
482 'skip': 'Blocked at Travis CI',
484 # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
485 'url': 'http://www.mtv.de/folgen/6b1ylu/teen-mom-2-enthuellungen-S5-F1',
487 'id': '1e5a878b-31c5-11e7-a442-0e40cf2fc285',
489 'title': 'Teen Mom 2',
490 'description': 'md5:dc65e357ef7e1085ed53e9e9d83146a7',
494 'skip_download': True,
496 'skip': 'Blocked at Travis CI',
498 'url': 'http://www.mtv.de/news/glolix/77491-mtv-movies-spotlight--pixels--teil-3',
500 'id': 'local_playlist-4e760566473c4c8c5344',
502 'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
503 'description': 'MTV Movies Supercut',
507 'skip_download': True,
509 'skip': 'Das Video kann zur Zeit nicht abgespielt werden.',
511 _GEO_COUNTRIES
= ['DE']
512 _FEED_URL
= 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
514 def _get_feed_query(self
, uri
):
521 class MTVItaliaIE(MTVServicesInfoExtractor
):
523 _VALID_URL
= r
'https?://(?:www\.)?mtv\.it/(?:episodi|video|musica)/(?P<id>[0-9a-z]+)'
525 'url': 'http://www.mtv.it/episodi/24bqab/mario-una-serie-di-maccio-capatonda-cavoli-amario-episodio-completo-S1-E1',
527 'id': '0f0fc78e-45fc-4cce-8f24-971c25477530',
529 'title': 'Cavoli amario (episodio completo)',
530 'description': 'md5:4962bccea8fed5b7c03b295ae1340660',
531 'series': 'Mario - Una Serie Di Maccio Capatonda',
536 'skip_download': True,
539 _GEO_COUNTRIES
= ['IT']
540 _FEED_URL
= 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
542 def _get_feed_query(self
, uri
):
549 class MTVItaliaProgrammaIE(MTVItaliaIE
):
550 IE_NAME
= 'mtv.it:programma'
551 _VALID_URL
= r
'https?://(?:www\.)?mtv\.it/(?:programmi|playlist)/(?P<id>[0-9a-z]+)'
553 # program page: general
554 'url': 'http://www.mtv.it/programmi/s2rppv/mario-una-serie-di-maccio-capatonda',
556 'id': 'a6f155bc-8220-4640-aa43-9b95f64ffa3d',
557 'title': 'Mario - Una Serie Di Maccio Capatonda',
558 'description': 'md5:72fbffe1f77ccf4e90757dd4e3216153',
562 'skip_download': True,
565 # program page: specific season
566 'url': 'http://www.mtv.it/programmi/d9ncjf/mario-una-serie-di-maccio-capatonda-S2',
568 'id': '4deeb5d8-f272-490c-bde2-ff8d261c6dd1',
569 'title': 'Mario - Una Serie Di Maccio Capatonda - Stagione 2',
571 'playlist_count': 34,
573 'skip_download': True,
576 # playlist page + redirect
577 'url': 'http://www.mtv.it/playlist/sexy-videos/ilctal',
579 'id': 'dee8f9ee-756d-493b-bf37-16d1d2783359',
580 'title': 'Sexy Videos',
582 'playlist_mincount': 145,
584 'skip_download': True,
587 _GEO_COUNTRIES
= ['IT']
588 _FEED_URL
= 'http://www.mtv.it/feeds/triforce/manifest/v8'
590 def _get_entries(self
, title
, url
):
592 pg
= self
._search
_regex
(r
'/(\d+)$', url
, 'entries', '1')
593 entries
= self
._download
_json
(url
, title
, 'page %s' % pg
)
595 entries
, lambda x
: x
['result']['nextPageURL'], compat_str
)
598 lambda x
: x
['result']['data']['items'],
599 lambda x
: x
['result']['data']['seasons']),
601 for entry
in entries
or []:
602 if entry
.get('canonicalURL'):
603 yield self
.url_result(entry
['canonicalURL'])
607 def _real_extract(self
, url
):
609 info_url
= update_url_query(self
._FEED
_URL
, query
)
610 video_id
= self
._match
_id
(url
)
611 info
= self
._download
_json
(info_url
, video_id
).get('manifest')
614 info
, lambda x
: x
['newLocation']['url'], compat_str
)
616 return self
.url_result(redirect
)
618 title
= info
.get('title')
620 info
, lambda x
: x
['reporting']['itemId'], compat_str
)
622 info
, lambda x
: x
['reporting']['parentId'], compat_str
)
624 playlist_url
= current_url
= None
625 for z
in (info
.get('zones') or {}).values():
626 if z
.get('moduleName') in ('INTL_M304', 'INTL_M209'):
627 info_url
= z
.get('feed')
628 if z
.get('moduleName') in ('INTL_M308', 'INTL_M317'):
629 playlist_url
= playlist_url
or z
.get('feed')
630 if z
.get('moduleName') in ('INTL_M300',):
631 current_url
= current_url
or z
.get('feed')
634 raise ExtractorError('No info found')
636 if video_id
== parent_id
:
637 video_id
= self
._search
_regex
(
638 r
'([^\/]+)/[^\/]+$', info_url
, 'video_id')
640 info
= self
._download
_json
(info_url
, video_id
, 'Show infos')
641 info
= try_get(info
, lambda x
: x
['result']['data'], dict)
642 title
= title
or try_get(
644 lambda x
: x
['title'],
645 lambda x
: x
['headline']),
647 description
= try_get(info
, lambda x
: x
['content'], compat_str
)
651 self
._download
_json
(playlist_url
, video_id
, 'Seasons info'),
652 lambda x
: x
['result']['data'], dict)
654 season
, lambda x
: x
['currentSeason'], compat_str
)
656 season
, lambda x
: x
['seasons'], list) or []
658 if current
in [s
.get('eTitle') for s
in seasons
]:
659 playlist_url
= current_url
662 r
'[-|]\s*(?:mtv\s*italia|programma|playlist)',
663 '', title
, flags
=re
.IGNORECASE
).strip()
665 return self
.playlist_result(
666 self
._get
_entries
(title
, playlist_url
),
667 video_id
, title
, description
)