from .common import InfoExtractor
from ..compat import (
- compat_urllib_parse_urlencode,
compat_str,
compat_xpath,
)
fix_xml_ampersands,
float_or_none,
HEADRequest,
+ NO_DEFAULT,
+ RegexNotFoundError,
sanitized_Request,
strip_or_none,
timeconvert,
unescapeHTML,
+ update_url_query,
url_basename,
- RegexNotFoundError,
xpath_text,
)
def _id_from_uri(uri):
return uri.split(':')[-1]
+ @staticmethod
+ def _remove_template_parameter(url):
+ # Remove the templates, like &device={device}
+ return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url)
+
# This was originally implemented for ComedyCentral, but it also works here
@classmethod
def _transform_rtmp_url(cls, rtmp_video_url):
url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
return [{'url': url, 'ext': 'mp4'}]
- def _extract_video_formats(self, mdoc, mtvn_id):
+ def _extract_video_formats(self, mdoc, mtvn_id, video_id):
if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', 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 '
formats = []
for rendition in mdoc.findall('.//rendition'):
- try:
- _, _, ext = rendition.attrib['type'].partition('/')
- rtmp_video_url = rendition.find('./src').text
- if rtmp_video_url.endswith('siteunavail.png'):
- continue
- new_urls = self._transform_rtmp_url(rtmp_video_url)
- formats.extend([{
- 'ext': 'flv' if new_url.startswith('rtmp') else ext,
- 'url': new_url,
- 'format_id': '-'.join(filter(None, [kind, rendition.get('bitrate')])),
- 'width': int(rendition.get('width')),
- 'height': int(rendition.get('height')),
- } for kind, new_url in new_urls.items()])
- except (KeyError, TypeError):
- raise ExtractorError('Invalid rendition field.')
+ if rendition.attrib['method'] == 'hls':
+ hls_url = rendition.find('./src').text
+ formats.extend(self._extract_m3u8_formats(hls_url, video_id, ext='mp4'))
+ else:
+ # fms
+ try:
+ _, _, ext = rendition.attrib['type'].partition('/')
+ rtmp_video_url = rendition.find('./src').text
+ if rtmp_video_url.endswith('siteunavail.png'):
+ continue
+ new_urls = self._transform_rtmp_url(rtmp_video_url)
+ formats.extend([{
+ 'ext': 'flv' if new_url.startswith('rtmp') else ext,
+ 'url': new_url,
+ 'format_id': '-'.join(filter(None, [kind, rendition.get('bitrate')])),
+ 'width': int(rendition.get('width')),
+ 'height': int(rendition.get('height')),
+ } for kind, new_url in new_urls.items()])
+ except (KeyError, TypeError):
+ raise ExtractorError('Invalid rendition field.')
self._sort_formats(formats)
return formats
} for typographic in transcript.findall('./typographic')]
return subtitles
- def _get_video_info(self, itemdoc):
+ def _get_video_info(self, itemdoc, use_hls):
uri = itemdoc.find('guid').text
video_id = self._id_from_uri(uri)
self.report_extraction(video_id)
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)
+ mediagen_url = self._remove_template_parameter(content_el.attrib['url'])
+ mediagen_url = mediagen_url.replace('device={device}', '')
if 'acceptMethods' not in mediagen_url:
mediagen_url += '&' if '?' in mediagen_url else '?'
- mediagen_url += 'acceptMethods=fms'
+ mediagen_url += 'acceptMethods='
+ mediagen_url += 'hls' if use_hls else 'fms'
mediagen_doc = self._download_xml(mediagen_url, video_id,
'Downloading video urls')
if mtvn_id_node is not None:
mtvn_id = mtvn_id_node.text
+ formats = self._extract_video_formats(mediagen_doc, mtvn_id, video_id)
+
return {
'title': title,
- 'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
+ 'formats': formats,
'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id),
'id': video_id,
'thumbnail': self._get_thumbnail_url(uri, itemdoc),
data = {'uri': uri}
if self._LANG:
data['lang'] = self._LANG
- return compat_urllib_parse_urlencode(data)
+ return data
- def _get_videos_info(self, uri):
+ def _get_videos_info(self, uri, use_hls=False):
video_id = self._id_from_uri(uri)
feed_url = self._get_feed_url(uri)
- info_url = feed_url + '?' + self._get_feed_query(uri)
- return self._get_videos_info_from_url(info_url, video_id)
+ info_url = update_url_query(feed_url, self._get_feed_query(uri))
+ return self._get_videos_info_from_url(info_url, video_id, use_hls)
- def _get_videos_info_from_url(self, url, video_id):
+ def _get_videos_info_from_url(self, url, video_id, use_hls):
idoc = self._download_xml(
url, video_id,
'Downloading info', transform_source=fix_xml_ampersands)
description = xpath_text(idoc, './channel/description')
return self.playlist_result(
- [self._get_video_info(item) for item in idoc.findall('.//item')],
+ [self._get_video_info(item, use_hls) for item in idoc.findall('.//item')],
playlist_title=title, playlist_description=description)
- def _extract_mgid(self, webpage):
+ def _extract_mgid(self, webpage, default=NO_DEFAULT):
try:
# the url can be http://media.mtvnservices.com/fb/{mgid}.swf
# or http://media.mtvnservices.com/{mgid}
sm4_embed = self._html_search_meta(
'sm4:video:embed', webpage, 'sm4 embed', default='')
mgid = self._search_regex(
- r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid')
+ r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid', default=default)
return mgid
def _real_extract(self, 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-arc/e1/players/{0}/'
- 'context52/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]
- return feed_url
+ config = self._download_json(
+ 'http://media.mtvnservices.com/pmt/e1/access/index.html?uri=%s&configtype=edge' % uri, video_id)
+ return self._remove_template_parameter(config['feedWithQueryParams'])
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
class MTVIE(MTVServicesInfoExtractor):
+ IE_NAME = 'mtv'
+ _VALID_URL = r'https?://(?:www\.)?mtv\.com/(?:video-clips|full-episodes)/(?P<id>[^/?#.]+)'
+ _FEED_URL = 'http://www.mtv.com/feeds/mrss/'
+
+ _TESTS = [{
+ 'url': 'http://www.mtv.com/video-clips/vl8qof/unlocking-the-truth-trailer',
+ 'md5': '1edbcdf1e7628e414a8c5dcebca3d32b',
+ 'info_dict': {
+ 'id': '5e14040d-18a4-47c4-a582-43ff602de88e',
+ 'ext': 'mp4',
+ 'title': 'Unlocking The Truth|July 18, 2016|1|101|Trailer',
+ 'description': '"Unlocking the Truth" premieres August 17th at 11/10c.',
+ 'timestamp': 1468846800,
+ 'upload_date': '20160718',
+ },
+ }, {
+ 'url': 'http://www.mtv.com/full-episodes/94tujl/unlocking-the-truth-gates-of-hell-season-1-ep-101',
+ 'only_matching': True,
+ }]
+
+
+class MTVVideoIE(MTVServicesInfoExtractor):
+ IE_NAME = 'mtv:video'
_VALID_URL = r'''(?x)^https?://
(?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''