youtube_dl/extractor/voxmedia.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from .once import OnceIE
   6 from ..compat import compat_urllib_parse_unquote
   7 from ..utils import ExtractorError
   8
   9
  10 class VoxMediaVolumeIE(OnceIE):
  11     _VALID_URL = r'https?://volume\.vox-cdn\.com/embed/(?P<id>[0-9a-f]{9})'
  12
  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:
  21                 continue
  22             info = {
  23                 'id': 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')
  27             }
  28             if provider_video_type == 'brightcove':
  29                 info['formats'] = self._extract_once_formats(provider_video_id)
  30                 self._sort_formats(info['formats'])
  31             else:
  32                 info.update({
  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(),
  36                 })
  37             return info
  38         raise ExtractorError('Unable to find provider video id')
  39
  40
  41 class VoxMediaIE(InfoExtractor):
  42     _VALID_URL = r'https?://(?:www\.)?(?:(?:theverge|vox|sbnation|eater|polygon|curbed|racked)\.com|recode\.net)/(?:[^/]+/)*(?P<id>[^/?]+)'
  43     _TESTS = [{
  44         'url': 'http://www.theverge.com/2014/6/27/5849272/material-world-how-google-discovered-what-software-is-made-of',
  45         'info_dict': {
  46             'id': '11eXZobjrG8DCSTgrNjVinU-YmmdYjhe',
  47             'ext': 'mp4',
  48             'title': 'Google\'s new material design direction',
  49             'description': 'md5:2f44f74c4d14a1f800ea73e1c6832ad2',
  50         },
  51         'params': {
  52             # m3u8 download
  53             'skip_download': True,
  54         },
  55         'add_ie': ['Ooyala'],
  56     }, {
  57         # data-ooyala-id
  58         'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet',
  59         'md5': 'd744484ff127884cd2ba09e3fa604e4b',
  60         'info_dict': {
  61             'id': 'RkZXU4cTphOCPDMZg5oEounJyoFI0g-B',
  62             'ext': 'mp4',
  63             'title': 'The Nexus 6: hands-on with Google\'s phablet',
  64             'description': 'md5:87a51fe95ff8cea8b5bdb9ac7ae6a6af',
  65         },
  66         'add_ie': ['Ooyala'],
  67         'skip': 'Video Not Found',
  68     }, {
  69         # volume embed
  70         'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill',
  71         'info_dict': {
  72             'id': 'wydzk3dDpmRz7PQoXRsTIX6XTkPjYL0b',
  73             'ext': 'mp4',
  74             'title': 'The new frontier of LGBTQ civil rights, explained',
  75             'description': 'md5:0dc58e94a465cbe91d02950f770eb93f',
  76         },
  77         'params': {
  78             # m3u8 download
  79             'skip_download': True,
  80         },
  81         'add_ie': ['Ooyala'],
  82     }, {
  83         # youtube embed
  84         'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance',
  85         'md5': '83b3080489fb103941e549352d3e0977',
  86         'info_dict': {
  87             'id': 'FcNHTJU1ufM',
  88             'ext': 'mp4',
  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',
  93             'uploader': 'Vox',
  94         },
  95         'add_ie': ['Youtube'],
  96     }, {
  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',
  99         'info_dict': {
 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!',
 103         },
 104         'playlist': [{
 105             'md5': '721fededf2ab74ae4176c8c8cbfe092e',
 106             'info_dict': {
 107                 'id': 'p3cThlMjE61VDi_SD9JlIteSNPWVDBB9',
 108                 'ext': 'mp4',
 109                 'title': 'Buddy Hield vs Steph Curry (and the world)',
 110                 'description': 'Let’s dissect only the most important Final Four storylines.',
 111             },
 112         }, {
 113             'md5': 'bf0c5cc115636af028be1bab79217ea9',
 114             'info_dict': {
 115                 'id': 'BmbmVjMjE6esPHxdALGubTrouQ0jYLHj',
 116                 'ext': 'mp4',
 117                 'title': 'Chasing Cinderella 2016: Syracuse basketball',
 118                 'description': 'md5:e02d56b026d51aa32c010676765a690d',
 119             },
 120         }],
 121     }, {
 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',
 125         'info_dict': {
 126             'id': '1231c973d',
 127             'ext': 'mp4',
 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.',
 130         },
 131         'add_ie': ['VoxMediaVolume'],
 132     }]
 133
 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))
 137
 138         def create_entry(provider_video_id, provider_video_type, title=None, description=None):
 139             video_url = {
 140                 'youtube': '%s',
 141                 'ooyala': 'ooyala:%s',
 142                 'volume': 'http://volume.vox-cdn.com/embed/%s',
 143             }[provider_video_type] % provider_video_id
 144             return {
 145                 '_type': 'url_transparent',
 146                 'url': video_url,
 147                 'title': title or self._og_search_title(webpage),
 148                 'description': description or self._og_search_description(webpage),
 149             }
 150
 151         entries = []
 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)
 157         if entries_data:
 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')))
 168
 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'))
 173
 174         volume_uuid = self._search_regex(
 175             r'data-volume-uuid="([^"]+)"', webpage, 'volume uuid', default=None)
 176         if volume_uuid:
 177             entries.append(create_entry(volume_uuid, 'volume'))
 178
 179         if len(entries) == 1:
 180             return entries[0]
 181         else:
 182             return self.playlist_result(entries, display_id, self._og_search_title(webpage), self._og_search_description(webpage))