voxmedia.py (9844B)
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 ( 8 ExtractorError, 9 int_or_none, 10 try_get, 11 unified_timestamp, 12 ) 13 14 15 class VoxMediaVolumeIE(OnceIE): 16 _VALID_URL = r'https?://volume\.vox-cdn\.com/embed/(?P<id>[0-9a-f]{9})' 17 18 def _real_extract(self, url): 19 video_id = self._match_id(url) 20 webpage = self._download_webpage(url, video_id) 21 22 setup = self._parse_json(self._search_regex( 23 r'setup\s*=\s*({.+});', webpage, 'setup'), video_id) 24 player_setup = setup.get('player_setup') or setup 25 video_data = player_setup.get('video') or {} 26 formatted_metadata = video_data.get('formatted_metadata') or {} 27 info = { 28 'id': video_id, 29 'title': player_setup.get('title') or video_data.get('title_short'), 30 'description': video_data.get('description_long') or video_data.get('description_short'), 31 'thumbnail': formatted_metadata.get('thumbnail') or video_data.get('brightcove_thumbnail'), 32 'timestamp': unified_timestamp(formatted_metadata.get('video_publish_date')), 33 } 34 asset = try_get(setup, lambda x: x['embed_assets']['chorus'], dict) or {} 35 36 formats = [] 37 hls_url = asset.get('hls_url') 38 if hls_url: 39 formats.extend(self._extract_m3u8_formats( 40 hls_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) 41 mp4_url = asset.get('mp4_url') 42 if mp4_url: 43 tbr = self._search_regex(r'-(\d+)k\.', mp4_url, 'bitrate', default=None) 44 format_id = 'http' 45 if tbr: 46 format_id += '-' + tbr 47 formats.append({ 48 'format_id': format_id, 49 'url': mp4_url, 50 'tbr': int_or_none(tbr), 51 }) 52 if formats: 53 self._sort_formats(formats) 54 info['formats'] = formats 55 info['duration'] = int_or_none(asset.get('duration')) 56 return info 57 58 for provider_video_type in ('ooyala', 'youtube', 'brightcove'): 59 provider_video_id = video_data.get('%s_id' % provider_video_type) 60 if not provider_video_id: 61 continue 62 if provider_video_type == 'brightcove': 63 info['formats'] = self._extract_once_formats(provider_video_id) 64 self._sort_formats(info['formats']) 65 else: 66 info.update({ 67 '_type': 'url_transparent', 68 'url': provider_video_id if provider_video_type == 'youtube' else '%s:%s' % (provider_video_type, provider_video_id), 69 'ie_key': provider_video_type.capitalize(), 70 }) 71 return info 72 raise ExtractorError('Unable to find provider video id') 73 74 75 class VoxMediaIE(InfoExtractor): 76 _VALID_URL = r'https?://(?:www\.)?(?:(?:theverge|vox|sbnation|eater|polygon|curbed|racked|funnyordie)\.com|recode\.net)/(?:[^/]+/)*(?P<id>[^/?]+)' 77 _TESTS = [{ 78 # Volume embed, Youtube 79 'url': 'http://www.theverge.com/2014/6/27/5849272/material-world-how-google-discovered-what-software-is-made-of', 80 'info_dict': { 81 'id': 'j4mLW6x17VM', 82 'ext': 'mp4', 83 'title': 'Material world: how Google discovered what software is made of', 84 'description': 'md5:dfc17e7715e3b542d66e33a109861382', 85 'upload_date': '20190710', 86 'uploader_id': 'TheVerge', 87 'uploader': 'The Verge', 88 }, 89 'add_ie': ['Youtube'], 90 }, { 91 # Volume embed, Youtube 92 'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet', 93 'md5': 'fd19aa0cf3a0eea515d4fd5c8c0e9d68', 94 'info_dict': { 95 'id': 'Gy8Md3Eky38', 96 'ext': 'mp4', 97 'title': 'The Nexus 6: hands-on with Google\'s phablet', 98 'description': 'md5:d9f0216e5fb932dd2033d6db37ac3f1d', 99 'uploader_id': 'TheVerge', 100 'upload_date': '20141021', 101 'uploader': 'The Verge', 102 'timestamp': 1413907200, 103 }, 104 'add_ie': ['Youtube'], 105 'skip': 'similar to the previous test', 106 }, { 107 # Volume embed, Youtube 108 'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill', 109 'info_dict': { 110 'id': '22986359b', 111 'ext': 'mp4', 112 'title': "Mississippi's laws are so bad that its anti-LGBTQ law isn't needed to allow discrimination", 113 'description': 'md5:fc1317922057de31cd74bce91eb1c66c', 114 'upload_date': '20150915', 115 'timestamp': 1442332800, 116 'duration': 285, 117 }, 118 'add_ie': ['Youtube'], 119 'skip': 'similar to the previous test', 120 }, { 121 # youtube embed 122 'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance', 123 'md5': '83b3080489fb103941e549352d3e0977', 124 'info_dict': { 125 'id': 'FcNHTJU1ufM', 126 'ext': 'mp4', 127 'title': 'How "the robot" became the greatest novelty dance of all time', 128 'description': 'md5:b081c0d588b8b2085870cda55e6da176', 129 'upload_date': '20160324', 130 'uploader_id': 'voxdotcom', 131 'uploader': 'Vox', 132 }, 133 'add_ie': ['Youtube'], 134 'skip': 'Page no longer contain videos', 135 }, { 136 # SBN.VideoLinkset.entryGroup multiple ooyala embeds 137 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok', 138 'info_dict': { 139 'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok', 140 'title': '25 lies you will tell yourself on National Signing Day', 141 'description': 'It\'s the most self-delusional time of the year, and everyone\'s gonna tell the same lies together!', 142 }, 143 'playlist': [{ 144 'md5': '721fededf2ab74ae4176c8c8cbfe092e', 145 'info_dict': { 146 'id': 'p3cThlMjE61VDi_SD9JlIteSNPWVDBB9', 147 'ext': 'mp4', 148 'title': 'Buddy Hield vs Steph Curry (and the world)', 149 'description': 'Let’s dissect only the most important Final Four storylines.', 150 }, 151 }, { 152 'md5': 'bf0c5cc115636af028be1bab79217ea9', 153 'info_dict': { 154 'id': 'BmbmVjMjE6esPHxdALGubTrouQ0jYLHj', 155 'ext': 'mp4', 156 'title': 'Chasing Cinderella 2016: Syracuse basketball', 157 'description': 'md5:e02d56b026d51aa32c010676765a690d', 158 }, 159 }], 160 'skip': 'Page no longer contain videos', 161 }, { 162 # volume embed, Brightcove Once 163 'url': 'https://www.recode.net/2014/6/17/11628066/post-post-pc-ceo-the-full-code-conference-video-of-microsofts-satya', 164 'md5': '2dbc77b8b0bff1894c2fce16eded637d', 165 'info_dict': { 166 'id': '1231c973d', 167 'ext': 'mp4', 168 'title': 'Post-Post-PC CEO: The Full Code Conference Video of Microsoft\'s Satya Nadella', 169 'description': 'The longtime veteran was chosen earlier this year as the software giant\'s third leader in its history.', 170 'timestamp': 1402938000, 171 'upload_date': '20140616', 172 'duration': 4114, 173 }, 174 'add_ie': ['VoxMediaVolume'], 175 }] 176 177 def _real_extract(self, url): 178 display_id = self._match_id(url) 179 webpage = compat_urllib_parse_unquote(self._download_webpage(url, display_id)) 180 181 def create_entry(provider_video_id, provider_video_type, title=None, description=None): 182 video_url = { 183 'youtube': '%s', 184 'ooyala': 'ooyala:%s', 185 'volume': 'http://volume.vox-cdn.com/embed/%s', 186 }[provider_video_type] % provider_video_id 187 return { 188 '_type': 'url_transparent', 189 'url': video_url, 190 'title': title or self._og_search_title(webpage), 191 'description': description or self._og_search_description(webpage), 192 } 193 194 entries = [] 195 entries_data = self._search_regex([ 196 r'Chorus\.VideoContext\.addVideo\((\[{.+}\])\);', 197 r'var\s+entry\s*=\s*({.+});', 198 r'SBN\.VideoLinkset\.entryGroup\(\s*(\[.+\])', 199 ], webpage, 'video data', default=None) 200 if entries_data: 201 entries_data = self._parse_json(entries_data, display_id) 202 if isinstance(entries_data, dict): 203 entries_data = [entries_data] 204 for video_data in entries_data: 205 provider_video_id = video_data.get('provider_video_id') 206 provider_video_type = video_data.get('provider_video_type') 207 if provider_video_id and provider_video_type: 208 entries.append(create_entry( 209 provider_video_id, provider_video_type, 210 video_data.get('title'), video_data.get('description'))) 211 212 provider_video_id = self._search_regex( 213 r'data-ooyala-id="([^"]+)"', webpage, 'ooyala id', default=None) 214 if provider_video_id: 215 entries.append(create_entry(provider_video_id, 'ooyala')) 216 217 volume_uuid = self._search_regex( 218 r'data-volume-uuid="([^"]+)"', webpage, 'volume uuid', default=None) 219 if volume_uuid: 220 entries.append(create_entry(volume_uuid, 'volume')) 221 222 if len(entries) == 1: 223 return entries[0] 224 else: 225 return self.playlist_result(entries, display_id, self._og_search_title(webpage), self._og_search_description(webpage))