github/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-07 20:30:41 -05:00
Code Issues Projects Releases Packages Wiki Activity

[comedycentral/mtv] Add support for HLS videos (fixes #11600)

Browse source 
Currently, the HTTP files of the RTMP urls are not present for the The Daily Show.
Use HLS instead for now.
This commit is contained in:
Philipp Hagemeister 2017-01-05 22:35:12 +01:00
parent a9ee260217
commit dbaf601646
2 changed files with 34 additions and 24 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
youtube_dl/extractor/comedycentral.py
View file

@ -57,7 +57,8 @@ def _real_extract(self, url):
feed = self._download_json(video_zone['feed'], playlist_id) feed = self._download_json(video_zone['feed'], playlist_id)
mgid = feed['result']['data']['id'] mgid = feed['result']['data']['id']
videos_info = self._get_videos_info(mgid) videos_info = self._get_videos_info(mgid, use_hls=True)
return videos_info return videos_info

55
youtube_dl/extractor/mtv.py
View file

@ -77,7 +77,7 @@ def _extract_mobile_video_formats(self, mtvn_id):
url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1) 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): 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.+?)?)$', mdoc.find('.//src').text) is not None: if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc.find('.//src').text) is not None:
if mtvn_id is not None and self._MOBILE_TEMPLATE is not None: if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
self.to_screen('The normal version is not available from your ' self.to_screen('The normal version is not available from your '
@ -88,21 +88,26 @@ def _extract_video_formats(self, mdoc, mtvn_id):
formats = [] formats = []
for rendition in mdoc.findall('.//rendition'): for rendition in mdoc.findall('.//rendition'):
try: if rendition.attrib['method'] == 'hls':
_, _, ext = rendition.attrib['type'].partition('/') hls_url = rendition.find('./src').text
rtmp_video_url = rendition.find('./src').text formats.extend(self._extract_m3u8_formats(hls_url, video_id, ext='mp4'))
if rtmp_video_url.endswith('siteunavail.png'): else:
continue # fms
new_urls = self._transform_rtmp_url(rtmp_video_url) try:
formats.extend([{ _, _, ext = rendition.attrib['type'].partition('/')
'ext': 'flv' if new_url.startswith('rtmp') else ext, rtmp_video_url = rendition.find('./src').text
'url': new_url, if rtmp_video_url.endswith('siteunavail.png'):
'format_id': '-'.join(filter(None, [kind, rendition.get('bitrate')])), continue
'width': int(rendition.get('width')), new_urls = self._transform_rtmp_url(rtmp_video_url)
'height': int(rendition.get('height')), formats.extend([{
} for kind, new_url in new_urls.items()]) 'ext': 'flv' if new_url.startswith('rtmp') else ext,
except (KeyError, TypeError): 'url': new_url,
raise ExtractorError('Invalid rendition field.') '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) self._sort_formats(formats)
return formats return formats
@ -118,15 +123,17 @@ def _extract_subtitles(self, mdoc, mtvn_id):
} for typographic in transcript.findall('./typographic')] } for typographic in transcript.findall('./typographic')]
return subtitles return subtitles
def _get_video_info(self, itemdoc): def _get_video_info(self, itemdoc, use_hls):
uri = itemdoc.find('guid').text uri = itemdoc.find('guid').text
video_id = self._id_from_uri(uri) video_id = self._id_from_uri(uri)
self.report_extraction(video_id) self.report_extraction(video_id)
content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))) content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content')))
mediagen_url = self._remove_template_parameter(content_el.attrib['url']) mediagen_url = self._remove_template_parameter(content_el.attrib['url'])
mediagen_url = mediagen_url.replace('device={device}', '')
if 'acceptMethods' not in mediagen_url: if 'acceptMethods' not in mediagen_url:
mediagen_url += '&' if '?' in mediagen_url else '?' 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, mediagen_doc = self._download_xml(mediagen_url, video_id,
'Downloading video urls') 'Downloading video urls')
@ -167,9 +174,11 @@ def _get_video_info(self, itemdoc):
if mtvn_id_node is not None: if mtvn_id_node is not None:
mtvn_id = mtvn_id_node.text mtvn_id = mtvn_id_node.text
formats = self._extract_video_formats(mediagen_doc, mtvn_id, video_id)
return { return {
'title': title, 'title': title,
'formats': self._extract_video_formats(mediagen_doc, mtvn_id), 'formats': formats,
'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id), 'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id),
'id': video_id, 'id': video_id,
'thumbnail': self._get_thumbnail_url(uri, itemdoc), 'thumbnail': self._get_thumbnail_url(uri, itemdoc),
@ -184,13 +193,13 @@ def _get_feed_query(self, uri):
data['lang'] = self._LANG data['lang'] = self._LANG
return 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) video_id = self._id_from_uri(uri)
feed_url = self._get_feed_url(uri) feed_url = self._get_feed_url(uri)
info_url = update_url_query(feed_url, self._get_feed_query(uri)) info_url = update_url_query(feed_url, self._get_feed_query(uri))
return self._get_videos_info_from_url(info_url, video_id) 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( idoc = self._download_xml(
url, video_id, url, video_id,
'Downloading info', transform_source=fix_xml_ampersands) 'Downloading info', transform_source=fix_xml_ampersands)
@ -199,7 +208,7 @@ def _get_videos_info_from_url(self, url, video_id):
description = xpath_text(idoc, './channel/description') description = xpath_text(idoc, './channel/description')
return self.playlist_result( 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) playlist_title=title, playlist_description=description)
def _extract_mgid(self, webpage, default=NO_DEFAULT): def _extract_mgid(self, webpage, default=NO_DEFAULT):