yt-dlp/youtube_dl/extractor/voxmedia.py

183 lines
7.9 KiB
Python
Raw Normal View History

# coding: utf-8
from __future__ import unicode_literals
from .common import InfoExtractor
from .once import OnceIE
from ..compat import compat_urllib_parse_unquote
from ..utils import ExtractorError
class VoxMediaVolumeIE(OnceIE):
_VALID_URL = r'https?://volume\.vox-cdn\.com/embed/(?P<id>[0-9a-f]{9})'
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
video_data = self._parse_json(self._search_regex(
r'Volume\.createVideo\(({.+})\s*,\s*{.*}\s*,\s*\[.*\]\s*,\s*{.*}\);', webpage, 'video data'), video_id)
for provider_video_type in ('ooyala', 'youtube', 'brightcove'):
provider_video_id = video_data.get('%s_id' % provider_video_type)
if not provider_video_id:
continue
info = {
'id': video_id,
'title': video_data.get('title_short'),
'description': video_data.get('description_long') or video_data.get('description_short'),
'thumbnail': video_data.get('brightcove_thumbnail')
}
if provider_video_type == 'brightcove':
info['formats'] = self._extract_once_formats(provider_video_id)
self._sort_formats(info['formats'])
else:
info.update({
'_type': 'url_transparent',
'url': provider_video_id if provider_video_type == 'youtube' else '%s:%s' % (provider_video_type, provider_video_id),
'ie_key': provider_video_type.capitalize(),
})
return info
raise ExtractorError('Unable to find provider video id')
class VoxMediaIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?(?:(?:theverge|vox|sbnation|eater|polygon|curbed|racked)\.com|recode\.net)/(?:[^/]+/)*(?P<id>[^/?]+)'
_TESTS = [{
'url': 'http://www.theverge.com/2014/6/27/5849272/material-world-how-google-discovered-what-software-is-made-of',
'info_dict': {
'id': '11eXZobjrG8DCSTgrNjVinU-YmmdYjhe',
'ext': 'mp4',
'title': 'Google\'s new material design direction',
'description': 'md5:2f44f74c4d14a1f800ea73e1c6832ad2',
},
'params': {
# m3u8 download
'skip_download': True,
},
'add_ie': ['Ooyala'],
}, {
# data-ooyala-id
'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet',
'md5': 'd744484ff127884cd2ba09e3fa604e4b',
'info_dict': {
'id': 'RkZXU4cTphOCPDMZg5oEounJyoFI0g-B',
'ext': 'mp4',
'title': 'The Nexus 6: hands-on with Google\'s phablet',
'description': 'md5:87a51fe95ff8cea8b5bdb9ac7ae6a6af',
},
'add_ie': ['Ooyala'],
'skip': 'Video Not Found',
}, {
# volume embed
'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill',
'info_dict': {
'id': 'wydzk3dDpmRz7PQoXRsTIX6XTkPjYL0b',
'ext': 'mp4',
'title': 'The new frontier of LGBTQ civil rights, explained',
'description': 'md5:0dc58e94a465cbe91d02950f770eb93f',
},
'params': {
# m3u8 download
'skip_download': True,
},
'add_ie': ['Ooyala'],
}, {
# youtube embed
'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance',
'md5': '83b3080489fb103941e549352d3e0977',
'info_dict': {
'id': 'FcNHTJU1ufM',
'ext': 'mp4',
'title': 'How "the robot" became the greatest novelty dance of all time',
'description': 'md5:b081c0d588b8b2085870cda55e6da176',
'upload_date': '20160324',
'uploader_id': 'voxdotcom',
'uploader': 'Vox',
},
'add_ie': ['Youtube'],
2016-04-01 00:33:36 +02:00
}, {
# SBN.VideoLinkset.entryGroup multiple ooyala embeds
'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
'info_dict': {
'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
'title': '25 lies you will tell yourself on National Signing Day',
'description': 'It\'s the most self-delusional time of the year, and everyone\'s gonna tell the same lies together!',
},
'playlist': [{
'md5': '721fededf2ab74ae4176c8c8cbfe092e',
'info_dict': {
'id': 'p3cThlMjE61VDi_SD9JlIteSNPWVDBB9',
'ext': 'mp4',
'title': 'Buddy Hield vs Steph Curry (and the world)',
'description': 'Lets dissect only the most important Final Four storylines.',
},
}, {
'md5': 'bf0c5cc115636af028be1bab79217ea9',
'info_dict': {
'id': 'BmbmVjMjE6esPHxdALGubTrouQ0jYLHj',
'ext': 'mp4',
'title': 'Chasing Cinderella 2016: Syracuse basketball',
'description': 'md5:e02d56b026d51aa32c010676765a690d',
},
}],
}, {
# volume embed, Brightcove Once
'url': 'https://www.recode.net/2014/6/17/11628066/post-post-pc-ceo-the-full-code-conference-video-of-microsofts-satya',
'md5': '01571a896281f77dc06e084138987ea2',
'info_dict': {
'id': '1231c973d',
'ext': 'mp4',
'title': 'Post-Post-PC CEO: The Full Code Conference Video of Microsoft\'s Satya Nadella',
'description': 'The longtime veteran was chosen earlier this year as the software giant\'s third leader in its history.',
},
'add_ie': ['VoxMediaVolume'],
}]
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = compat_urllib_parse_unquote(self._download_webpage(url, display_id))
2016-04-01 00:33:36 +02:00
def create_entry(provider_video_id, provider_video_type, title=None, description=None):
video_url = {
'youtube': '%s',
'ooyala': 'ooyala:%s',
'volume': 'http://volume.vox-cdn.com/embed/%s',
}[provider_video_type] % provider_video_id
2016-04-01 00:33:36 +02:00
return {
'_type': 'url_transparent',
'url': video_url,
2016-04-01 00:33:36 +02:00
'title': title or self._og_search_title(webpage),
'description': description or self._og_search_description(webpage),
}
2016-04-01 00:33:36 +02:00
entries = []
entries_data = self._search_regex([
r'Chorus\.VideoContext\.addVideo\((\[{.+}\])\);',
r'var\s+entry\s*=\s*({.+});',
r'SBN\.VideoLinkset\.entryGroup\(\s*(\[.+\])',
], webpage, 'video data', default=None)
2016-04-01 00:33:36 +02:00
if entries_data:
entries_data = self._parse_json(entries_data, display_id)
if isinstance(entries_data, dict):
entries_data = [entries_data]
for video_data in entries_data:
provider_video_id = video_data.get('provider_video_id')
provider_video_type = video_data.get('provider_video_type')
if provider_video_id and provider_video_type:
entries.append(create_entry(
provider_video_id, provider_video_type,
video_data.get('title'), video_data.get('description')))
2016-04-01 00:33:36 +02:00
provider_video_id = self._search_regex(
r'data-ooyala-id="([^"]+)"', webpage, 'ooyala id', default=None)
if provider_video_id:
entries.append(create_entry(provider_video_id, 'ooyala'))
2016-04-01 00:33:36 +02:00
volume_uuid = self._search_regex(
r'data-volume-uuid="([^"]+)"', webpage, 'volume uuid', default=None)
if volume_uuid:
entries.append(create_entry(volume_uuid, 'volume'))
2016-04-01 00:33:36 +02:00
if len(entries) == 1:
return entries[0]
else:
return self.playlist_result(entries, display_id, self._og_search_title(webpage), self._og_search_description(webpage))