2 from __future__
import unicode_literals
4 from .common
import InfoExtractor
5 from .once
import OnceIE
6 from ..compat
import compat_urllib_parse_unquote
7 from ..utils
import ExtractorError
10 class VoxMediaVolumeIE(OnceIE
):
11 _VALID_URL
= r
'https?://volume\.vox-cdn\.com/embed/(?P<id>[0-9a-f]{9})'
13 def _real_extract(self
, url
):
14 video_id
= self
._match
_id
(url
)
15 webpage
= self
._download
_webpage
(url
, video_id
)
16 video_data
= self
._parse
_json
(self
._search
_regex
(
17 r
'Volume\.createVideo\(({.+})\s*,\s*{.*}\s*,\s*\[.*\]\s*,\s*{.*}\);', webpage
, 'video data'), video_id
)
18 for provider_video_type
in ('ooyala', 'youtube', 'brightcove'):
19 provider_video_id
= video_data
.get('%s_id' % provider_video_type
)
20 if not provider_video_id
:
24 'title': video_data
.get('title_short'),
25 'description': video_data
.get('description_long') or video_data
.get('description_short'),
26 'thumbnail': video_data
.get('brightcove_thumbnail')
28 if provider_video_type
== 'brightcove':
29 info
['formats'] = self
._extract
_once
_formats
(provider_video_id
)
30 self
._sort
_formats
(info
['formats'])
33 '_type': 'url_transparent',
34 'url': provider_video_id
if provider_video_type
== 'youtube' else '%s:%s' % (provider_video_type
, provider_video_id
),
35 'ie_key': provider_video_type
.capitalize(),
38 raise ExtractorError('Unable to find provider video id')
41 class VoxMediaIE(InfoExtractor
):
42 _VALID_URL
= r
'https?://(?:www\.)?(?:(?:theverge|vox|sbnation|eater|polygon|curbed|racked)\.com|recode\.net)/(?:[^/]+/)*(?P<id>[^/?]+)'
44 'url': 'http://www.theverge.com/2014/6/27/5849272/material-world-how-google-discovered-what-software-is-made-of',
46 'id': '11eXZobjrG8DCSTgrNjVinU-YmmdYjhe',
48 'title': 'Google\'s new material design direction',
49 'description': 'md5:2f44f74c4d14a1f800ea73e1c6832ad2',
53 'skip_download': True,
58 'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet',
59 'md5': 'd744484ff127884cd2ba09e3fa604e4b',
61 'id': 'RkZXU4cTphOCPDMZg5oEounJyoFI0g-B',
63 'title': 'The Nexus 6: hands-on with Google\'s phablet',
64 'description': 'md5:87a51fe95ff8cea8b5bdb9ac7ae6a6af',
67 'skip': 'Video Not Found',
70 'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill',
72 'id': 'wydzk3dDpmRz7PQoXRsTIX6XTkPjYL0b',
74 'title': 'The new frontier of LGBTQ civil rights, explained',
75 'description': 'md5:0dc58e94a465cbe91d02950f770eb93f',
79 'skip_download': True,
84 'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance',
85 'md5': '83b3080489fb103941e549352d3e0977',
89 'title': 'How "the robot" became the greatest novelty dance of all time',
90 'description': 'md5:b081c0d588b8b2085870cda55e6da176',
91 'upload_date': '20160324',
92 'uploader_id': 'voxdotcom',
95 'add_ie': ['Youtube'],
97 # SBN.VideoLinkset.entryGroup multiple ooyala embeds
98 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
100 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
101 'title': '25 lies you will tell yourself on National Signing Day',
102 'description': 'It\'s the most self-delusional time of the year, and everyone\'s gonna tell the same lies together!',
105 'md5': '721fededf2ab74ae4176c8c8cbfe092e',
107 'id': 'p3cThlMjE61VDi_SD9JlIteSNPWVDBB9',
109 'title': 'Buddy Hield vs Steph Curry (and the world)',
110 'description': 'Let’s dissect only the most important Final Four storylines.',
113 'md5': 'bf0c5cc115636af028be1bab79217ea9',
115 'id': 'BmbmVjMjE6esPHxdALGubTrouQ0jYLHj',
117 'title': 'Chasing Cinderella 2016: Syracuse basketball',
118 'description': 'md5:e02d56b026d51aa32c010676765a690d',
122 # volume embed, Brightcove Once
123 'url': 'https://www.recode.net/2014/6/17/11628066/post-post-pc-ceo-the-full-code-conference-video-of-microsofts-satya',
124 'md5': '01571a896281f77dc06e084138987ea2',
128 'title': 'Post-Post-PC CEO: The Full Code Conference Video of Microsoft\'s Satya Nadella',
129 'description': 'The longtime veteran was chosen earlier this year as the software giant\'s third leader in its history.',
131 'add_ie': ['VoxMediaVolume'],
134 def _real_extract(self
, url
):
135 display_id
= self
._match
_id
(url
)
136 webpage
= compat_urllib_parse_unquote(self
._download
_webpage
(url
, display_id
))
138 def create_entry(provider_video_id
, provider_video_type
, title
=None, description
=None):
141 'ooyala': 'ooyala:%s',
142 'volume': 'http://volume.vox-cdn.com/embed/%s',
143 }[provider_video_type
] % provider_video_id
145 '_type': 'url_transparent',
147 'title': title
or self
._og
_search
_title
(webpage
),
148 'description': description
or self
._og
_search
_description
(webpage
),
152 entries_data
= self
._search
_regex
([
153 r
'Chorus\.VideoContext\.addVideo\((\[{.+}\])\);',
154 r
'var\s+entry\s*=\s*({.+});',
155 r
'SBN\.VideoLinkset\.entryGroup\(\s*(\[.+\])',
156 ], webpage
, 'video data', default
=None)
158 entries_data
= self
._parse
_json
(entries_data
, display_id
)
159 if isinstance(entries_data
, dict):
160 entries_data
= [entries_data
]
161 for video_data
in entries_data
:
162 provider_video_id
= video_data
.get('provider_video_id')
163 provider_video_type
= video_data
.get('provider_video_type')
164 if provider_video_id
and provider_video_type
:
165 entries
.append(create_entry(
166 provider_video_id
, provider_video_type
,
167 video_data
.get('title'), video_data
.get('description')))
169 provider_video_id
= self
._search
_regex
(
170 r
'data-ooyala-id="([^"]+)"', webpage
, 'ooyala id', default
=None)
171 if provider_video_id
:
172 entries
.append(create_entry(provider_video_id
, 'ooyala'))
174 volume_uuid
= self
._search
_regex
(
175 r
'data-volume-uuid="([^"]+)"', webpage
, 'volume uuid', default
=None)
177 entries
.append(create_entry(volume_uuid
, 'volume'))
179 if len(entries
) == 1:
182 return self
.playlist_result(entries
, display_id
, self
._og
_search
_title
(webpage
), self
._og
_search
_description
(webpage
))