import re
from .common import InfoExtractor
+from ..compat import (
+ compat_urllib_parse_urlencode,
+ compat_str,
+ compat_xpath,
+)
from ..utils import (
- compat_urllib_parse,
- compat_urllib_request,
ExtractorError,
find_xpath_attr,
fix_xml_ampersands,
+ float_or_none,
HEADRequest,
+ sanitized_Request,
unescapeHTML,
url_basename,
RegexNotFoundError,
+ xpath_text,
)
class MTVServicesInfoExtractor(InfoExtractor):
_MOBILE_TEMPLATE = None
+ _LANG = None
@staticmethod
def _id_from_uri(uri):
m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
if not m:
return rtmp_video_url
- base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
+ base = 'http://viacommtvstrmfs.fplive.net/'
return base + m.group('finalid')
def _get_feed_url(self, uri):
def _extract_mobile_video_formats(self, mtvn_id):
webpage_url = self._MOBILE_TEMPLATE % mtvn_id
- req = compat_urllib_request.Request(webpage_url)
+ req = sanitized_Request(webpage_url)
# Otherwise we get a webpage that would execute some javascript
- req.add_header('Youtubedl-user-agent', 'curl/7')
+ req.add_header('User-Agent', 'curl/7')
webpage = self._download_webpage(req, mtvn_id,
- 'Downloading mobile page')
+ 'Downloading mobile page')
metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
req = HEADRequest(metrics_url)
response = self._request_webpage(req, mtvn_id, 'Resolving url')
url = response.geturl()
# Transform the url to get the best quality:
url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
- return [{'url': url,'ext': 'mp4'}]
+ return [{'url': url, 'ext': 'mp4'}]
def _extract_video_formats(self, mdoc, mtvn_id):
- if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None:
+ if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc.find('.//src').text) is not None:
if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
self.to_screen('The normal version is not available from your '
- 'country, trying with the mobile version')
+ 'country, trying with the mobile version')
return self._extract_mobile_video_formats(mtvn_id)
raise ExtractorError('This video is not available from your country.',
- expected=True)
+ expected=True)
formats = []
for rendition in mdoc.findall('.//rendition'):
try:
_, _, ext = rendition.attrib['type'].partition('/')
rtmp_video_url = rendition.find('./src').text
- formats.append({'ext': ext,
- 'url': self._transform_rtmp_url(rtmp_video_url),
- 'format_id': rendition.get('bitrate'),
- 'width': int(rendition.get('width')),
- 'height': int(rendition.get('height')),
- })
+ if rtmp_video_url.endswith('siteunavail.png'):
+ continue
+ new_url = self._transform_rtmp_url(rtmp_video_url)
+ formats.append({
+ 'ext': 'flv' if new_url.startswith('rtmp') else ext,
+ 'url': new_url,
+ 'format_id': rendition.get('bitrate'),
+ 'width': int(rendition.get('width')),
+ 'height': int(rendition.get('height')),
+ })
except (KeyError, TypeError):
raise ExtractorError('Invalid rendition field.')
self._sort_formats(formats)
return formats
+ def _extract_subtitles(self, mdoc, mtvn_id):
+ subtitles = {}
+ for transcript in mdoc.findall('.//transcript'):
+ if transcript.get('kind') != 'captions':
+ continue
+ lang = transcript.get('srclang')
+ subtitles[lang] = [{
+ 'url': compat_str(typographic.get('src')),
+ 'ext': typographic.get('format')
+ } for typographic in transcript.findall('./typographic')]
+ return subtitles
+
def _get_video_info(self, itemdoc):
uri = itemdoc.find('guid').text
video_id = self._id_from_uri(uri)
self.report_extraction(video_id)
- mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
+ content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content')))
+ mediagen_url = content_el.attrib['url']
# Remove the templates, like &device={device}
mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
if 'acceptMethods' not in mediagen_url:
- mediagen_url += '&acceptMethods=fms'
+ mediagen_url += '&' if '?' in mediagen_url else '?'
+ mediagen_url += 'acceptMethods=fms'
mediagen_doc = self._download_xml(mediagen_url, video_id,
- 'Downloading video urls')
+ 'Downloading video urls')
- description_node = itemdoc.find('description')
- if description_node is not None:
- description = description_node.text.strip()
- else:
- description = None
+ item = mediagen_doc.find('./video/item')
+ if item is not None and item.get('type') == 'text':
+ message = '%s returned error: ' % self.IE_NAME
+ if item.get('code') is not None:
+ message += '%s - ' % item.get('code')
+ message += item.text
+ raise ExtractorError(message, expected=True)
+
+ description = xpath_text(itemdoc, 'description')
title_el = None
if title_el is None:
itemdoc, './/{http://search.yahoo.com/mrss/}category',
'scheme', 'urn:mtvn:video_title')
if title_el is None:
- title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
+ title_el = itemdoc.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title'))
if title_el is None:
- title_el = itemdoc.find('.//title')
+ title_el = itemdoc.find(compat_xpath('.//title'))
if title_el.text is None:
title_el = None
# This a short id that's used in the webpage urls
mtvn_id = None
mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
- 'scheme', 'urn:mtvn:id')
+ 'scheme', 'urn:mtvn:id')
if mtvn_id_node is not None:
mtvn_id = mtvn_id_node.text
return {
'title': title,
'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
+ 'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id),
'id': video_id,
'thumbnail': self._get_thumbnail_url(uri, itemdoc),
'description': description,
+ 'duration': float_or_none(content_el.attrib.get('duration')),
}
+ def _get_feed_query(self, uri):
+ data = {'uri': uri}
+ if self._LANG:
+ data['lang'] = self._LANG
+ return compat_urllib_parse_urlencode(data)
+
def _get_videos_info(self, uri):
video_id = self._id_from_uri(uri)
feed_url = self._get_feed_url(uri)
- data = compat_urllib_parse.urlencode({'uri': uri})
+ info_url = feed_url + '?' + self._get_feed_query(uri)
+ return self._get_videos_info_from_url(info_url, video_id)
+
+ def _get_videos_info_from_url(self, url, video_id):
idoc = self._download_xml(
- feed_url + '?' + data, video_id,
+ url, video_id,
'Downloading info', transform_source=fix_xml_ampersands)
- return [self._get_video_info(item) for item in idoc.findall('.//item')]
+ return self.playlist_result(
+ [self._get_video_info(item) for item in idoc.findall('.//item')])
- def _real_extract(self, url):
- title = url_basename(url)
- webpage = self._download_webpage(url, title)
+ def _extract_mgid(self, webpage):
try:
# the url can be http://media.mtvnservices.com/fb/{mgid}.swf
# or http://media.mtvnservices.com/{mgid}
if mgid.endswith('.swf'):
mgid = mgid[:-4]
except RegexNotFoundError:
+ mgid = None
+
+ if mgid is None or ':' not in mgid:
mgid = self._search_regex(
[r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
- webpage, u'mgid')
- return self._get_videos_info(mgid)
+ webpage, 'mgid', default=None)
+
+ if not mgid:
+ sm4_embed = self._html_search_meta(
+ 'sm4:video:embed', webpage, 'sm4 embed', default='')
+ mgid = self._search_regex(
+ r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid')
+ return mgid
+
+ def _real_extract(self, url):
+ title = url_basename(url)
+ webpage = self._download_webpage(url, title)
+ mgid = self._extract_mgid(webpage)
+ videos_info = self._get_videos_info(mgid)
+ return videos_info
class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
},
}
+ @staticmethod
+ def _extract_url(webpage):
+ mobj = re.search(
+ r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage)
+ if mobj:
+ return mobj.group('url')
+
def _get_feed_url(self, uri):
video_id = self._id_from_uri(uri)
site_id = uri.replace(video_id, '')
- config_url = 'http://media.mtvnservices.com/pmt/e1/players/{0}/config.xml'.format(site_id)
+ config_url = ('http://media.mtvnservices.com/pmt/e1/players/{0}/'
+ 'context4/context5/config.xml'.format(site_id))
config_doc = self._download_xml(config_url, video_id)
feed_node = config_doc.find('.//feed')
feed_url = feed_node.text.strip().split('?')[0]
_TESTS = [
{
'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
- 'file': '853555.mp4',
'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
'info_dict': {
+ 'id': '853555',
+ 'ext': 'mp4',
'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
},
},
- {
- 'add_ie': ['Vevo'],
- 'url': 'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
- 'file': 'USCJY1331283.mp4',
- 'md5': '73b4e7fcadd88929292fe52c3ced8caf',
- 'info_dict': {
- 'title': 'Everything Has Changed',
- 'upload_date': '20130606',
- 'uploader': 'Taylor Swift',
- },
- 'skip': 'VEVO is only available in some countries',
- },
]
def _get_thumbnail_url(self, uri, itemdoc):
uri = mobj.groupdict().get('mgid')
if uri is None:
webpage = self._download_webpage(url, video_id)
-
+
# Some videos come from Vevo.com
- m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
- webpage, re.DOTALL)
+ m_vevo = re.search(
+ r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage)
if m_vevo:
- vevo_id = m_vevo.group(1);
+ vevo_id = m_vevo.group(1)
self.to_screen('Vevo video detected: %s' % vevo_id)
return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
-
+
uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
return self._get_videos_info(uri)
}
}
_FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'
+
+
+class MTVDEIE(MTVServicesInfoExtractor):
+ IE_NAME = 'mtv.de'
+ _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$'
+ _TESTS = [{
+ 'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
+ 'info_dict': {
+ 'id': 'music_video-a50bc5f0b3aa4b3190aa',
+ 'ext': 'mp4',
+ 'title': 'MusicVideo_cro-traum',
+ 'description': 'Cro - Traum',
+ },
+ 'params': {
+ # rtmp download
+ 'skip_download': True,
+ },
+ }, {
+ # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
+ 'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen',
+ 'info_dict': {
+ 'id': 'local_playlist-f5ae778b9832cc837189',
+ 'ext': 'mp4',
+ 'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
+ },
+ 'params': {
+ # rtmp download
+ 'skip_download': True,
+ },
+ }, {
+ # single video in pagePlaylist with different id
+ 'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
+ 'info_dict': {
+ 'id': 'local_playlist-4e760566473c4c8c5344',
+ 'ext': 'mp4',
+ 'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
+ 'description': 'MTV Movies Supercut',
+ },
+ 'params': {
+ # rtmp download
+ 'skip_download': True,
+ },
+ }]
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+
+ webpage = self._download_webpage(url, video_id)
+
+ playlist = self._parse_json(
+ self._search_regex(
+ r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'),
+ video_id)
+
+ # news pages contain single video in playlist with different id
+ if len(playlist) == 1:
+ return self._get_videos_info_from_url(playlist[0]['mrss'], video_id)
+
+ for item in playlist:
+ item_id = item.get('id')
+ if item_id and compat_str(item_id) == video_id:
+ return self._get_videos_info_from_url(item['mrss'], video_id)