Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vevo.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_etree_fromstring
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     sanitized_Request,
  11     parse_iso8601,
  12 )
  13
  14
  15 class VevoIE(InfoExtractor):
  16     '''
  17     Accepts urls from vevo.com or in the format 'vevo:{id}'
  18     (currently used by MTVIE and MySpaceIE)
  19     '''
  20     _VALID_URL = r'''(?x)
  21         (?:https?://www\.vevo\.com/watch/(?:[^/]+/(?:[^/]+/)?)?|
  22            https?://cache\.vevo\.com/m/html/embed\.html\?video=|
  23            https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
  24            vevo:)
  25         (?P<id>[^&?#]+)'''
  26
  27     _TESTS = [{
  28         'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
  29         'md5': '95ee28ee45e70130e3ab02b0f579ae23',
  30         'info_dict': {
  31             'id': 'GB1101300280',
  32             'ext': 'mp4',
  33             'title': 'Somebody to Die For',
  34             'upload_date': '20130624',
  35             'uploader': 'Hurts',
  36             'timestamp': 1372057200,
  37         },
  38     }, {
  39         'note': 'v3 SMIL format',
  40         'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
  41         'md5': 'f6ab09b034f8c22969020b042e5ac7fc',
  42         'info_dict': {
  43             'id': 'USUV71302923',
  44             'ext': 'mp4',
  45             'title': 'I Wish I Could Break Your Heart',
  46             'upload_date': '20140219',
  47             'uploader': 'Cassadee Pope',
  48             'timestamp': 1392796919,
  49         },
  50     }, {
  51         'note': 'Age-limited video',
  52         'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
  53         'info_dict': {
  54             'id': 'USRV81300282',
  55             'ext': 'mp4',
  56             'title': 'Tunnel Vision (Explicit)',
  57             'upload_date': '20130703',
  58             'age_limit': 18,
  59             'uploader': 'Justin Timberlake',
  60             'timestamp': 1372888800,
  61         },
  62     }, {
  63         'note': 'No video_info',
  64         'url': 'http://www.vevo.com/watch/k-camp-1/Till-I-Die/USUV71503000',
  65         'md5': '8b83cc492d72fc9cf74a02acee7dc1b0',
  66         'info_dict': {
  67             'id': 'USUV71503000',
  68             'ext': 'mp4',
  69             'title': 'Till I Die',
  70             'upload_date': '20151207',
  71             'age_limit': 18,
  72             'uploader': 'K Camp',
  73             'timestamp': 1449468000,
  74         },
  75     }]
  76     _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com'
  77     _SOURCE_TYPES = {
  78         0: 'youtube',
  79         1: 'brightcove',
  80         2: 'http',
  81         3: 'hls_ios',
  82         4: 'hls',
  83         5: 'smil',  # http
  84         7: 'f4m_cc',
  85         8: 'f4m_ak',
  86         9: 'f4m_l3',
  87         10: 'ism',
  88         13: 'smil',  # rtmp
  89         18: 'dash',
  90     }
  91     _VERSIONS = {
  92         0: 'youtube',  # only in AuthenticateVideo videoVersions
  93         1: 'level3',
  94         2: 'akamai',
  95         3: 'level3',
  96         4: 'amazon',
  97     }
  98
  99     def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
 100         formats = []
 101         els = smil.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
 102         for el in els:
 103             src = el.attrib['src']
 104             m = re.match(r'''(?xi)
 105                 (?P<ext>[a-z0-9]+):
 106                 (?P<path>
 107                     [/a-z0-9]+     # The directory and main part of the URL
 108                     _(?P<tbr>[0-9]+)k
 109                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 110                     _(?P<vcodec>[a-z0-9]+)
 111                     _(?P<vbr>[0-9]+)
 112                     _(?P<acodec>[a-z0-9]+)
 113                     _(?P<abr>[0-9]+)
 114                     \.[a-z0-9]+  # File extension
 115                 )''', src)
 116             if not m:
 117                 continue
 118
 119             format_url = self._SMIL_BASE_URL + m.group('path')
 120             formats.append({
 121                 'url': format_url,
 122                 'format_id': 'smil_' + m.group('tbr'),
 123                 'vcodec': m.group('vcodec'),
 124                 'acodec': m.group('acodec'),
 125                 'tbr': int(m.group('tbr')),
 126                 'vbr': int(m.group('vbr')),
 127                 'abr': int(m.group('abr')),
 128                 'ext': m.group('ext'),
 129                 'width': int(m.group('width')),
 130                 'height': int(m.group('height')),
 131             })
 132         return formats
 133
 134     def _initialize_api(self, video_id):
 135         req = sanitized_Request(
 136             'http://www.vevo.com/auth', data=b'')
 137         webpage = self._download_webpage(
 138             req, None,
 139             note='Retrieving oauth token',
 140             errnote='Unable to retrieve oauth token')
 141
 142         if 'THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION' in webpage:
 143             raise ExtractorError(
 144                 '%s said: This page is currently unavailable in your region.' % self.IE_NAME, expected=True)
 145
 146         auth_info = self._parse_json(webpage, video_id)
 147         self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['access_token']
 148
 149     def _call_api(self, path, video_id, note, errnote, fatal=True):
 150         return self._download_json(self._api_url_template % path, video_id, note, errnote)
 151
 152     def _real_extract(self, url):
 153         video_id = self._match_id(url)
 154
 155         json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
 156         response = self._download_json(
 157             json_url, video_id, 'Downloading video info', 'Unable to download info')
 158         video_info = response.get('video') or {}
 159         video_versions = video_info.get('videoVersions')
 160         uploader = None
 161         timestamp = None
 162         view_count = None
 163         formats = []
 164
 165         if not video_info:
 166             if response.get('statusCode') != 909:
 167                 ytid = response.get('errorInfo', {}).get('ytid')
 168                 if ytid:
 169                     self.report_warning(
 170                         'Video is geoblocked, trying with the YouTube video %s' % ytid)
 171                     return self.url_result(ytid, 'Youtube', ytid)
 172
 173                 if 'statusMessage' in response:
 174                     raise ExtractorError('%s said: %s' % (
 175                         self.IE_NAME, response['statusMessage']), expected=True)
 176                 raise ExtractorError('Unable to extract videos')
 177
 178             self._initialize_api(video_id)
 179             video_info = self._call_api(
 180                 'video/%s' % video_id, video_id, 'Downloading api video info',
 181                 'Failed to download video info')
 182
 183             video_versions = self._call_api(
 184                 'video/%s/streams' % video_id, video_id,
 185                 'Downloading video versions info',
 186                 'Failed to download video versions info')
 187
 188             timestamp = parse_iso8601(video_info.get('releaseDate'))
 189             artists = video_info.get('artists')
 190             if artists:
 191                 uploader = artists[0]['name']
 192             view_count = int_or_none(video_info.get('views', {}).get('total'))
 193
 194             for video_version in video_versions:
 195                 version = self._VERSIONS.get(video_version['version'])
 196                 version_url = video_version.get('url')
 197                 if not version_url:
 198                     continue
 199
 200                 if '.ism' in version_url:
 201                     continue
 202                 elif '.mpd' in version_url:
 203                     formats.extend(self._extract_mpd_formats(
 204                         version_url, video_id, mpd_id='dash-%s' % version,
 205                         note='Downloading %s MPD information' % version,
 206                         errnote='Failed to download %s MPD information' % version,
 207                         fatal=False))
 208                 elif '.m3u8' in version_url:
 209                     formats.extend(self._extract_m3u8_formats(
 210                         version_url, video_id, 'mp4', 'm3u8_native',
 211                         m3u8_id='hls-%s' % version,
 212                         note='Downloading %s m3u8 information' % version,
 213                         errnote='Failed to download %s m3u8 information' % version,
 214                         fatal=False))
 215                 else:
 216                     m = re.search(r'''(?xi)
 217                         _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 218                         _(?P<vcodec>[a-z0-9]+)
 219                         _(?P<vbr>[0-9]+)
 220                         _(?P<acodec>[a-z0-9]+)
 221                         _(?P<abr>[0-9]+)
 222                         \.(?P<ext>[a-z0-9]+)''', version_url)
 223                     if not m:
 224                         continue
 225
 226                     formats.append({
 227                         'url': version_url,
 228                         'format_id': 'http-%s-%s' % (version, video_version['quality']),
 229                         'vcodec': m.group('vcodec'),
 230                         'acodec': m.group('acodec'),
 231                         'vbr': int(m.group('vbr')),
 232                         'abr': int(m.group('abr')),
 233                         'ext': m.group('ext'),
 234                         'width': int(m.group('width')),
 235                         'height': int(m.group('height')),
 236                     })
 237         else:
 238             timestamp = int_or_none(self._search_regex(
 239                 r'/Date\((\d+)\)/',
 240                 video_info['releaseDate'], 'release date', fatal=False),
 241                 scale=1000)
 242             artists = video_info.get('mainArtists')
 243             if artists:
 244                 uploader = artists[0]['artistName']
 245
 246             smil_parsed = False
 247             for video_version in video_info['videoVersions']:
 248                 version = self._VERSIONS.get(video_version['version'])
 249                 if version == 'youtube':
 250                     continue
 251                 else:
 252                     source_type = self._SOURCE_TYPES.get(video_version['sourceType'])
 253                     renditions = compat_etree_fromstring(video_version['data'])
 254                     if source_type == 'http':
 255                         for rend in renditions.findall('rendition'):
 256                             attr = rend.attrib
 257                             formats.append({
 258                                 'url': attr['url'],
 259                                 'format_id': 'http-%s-%s' % (version, attr['name']),
 260                                 'height': int_or_none(attr.get('frameheight')),
 261                                 'width': int_or_none(attr.get('frameWidth')),
 262                                 'tbr': int_or_none(attr.get('totalBitrate')),
 263                                 'vbr': int_or_none(attr.get('videoBitrate')),
 264                                 'abr': int_or_none(attr.get('audioBitrate')),
 265                                 'vcodec': attr.get('videoCodec'),
 266                                 'acodec': attr.get('audioCodec'),
 267                             })
 268                     elif source_type == 'hls':
 269                         formats.extend(self._extract_m3u8_formats(
 270                             renditions.find('rendition').attrib['url'], video_id,
 271                             'mp4', 'm3u8_native', m3u8_id='hls-%s' % version,
 272                             note='Downloading %s m3u8 information' % version,
 273                             errnote='Failed to download %s m3u8 information' % version,
 274                             fatal=False))
 275                     elif source_type == 'smil' and version == 'level3' and not smil_parsed:
 276                         formats.extend(self._extract_smil_formats(
 277                             renditions.find('rendition').attrib['url'], video_id, False))
 278                         smil_parsed = True
 279         self._sort_formats(formats)
 280
 281         title = video_info['title']
 282
 283         is_explicit = video_info.get('isExplicit')
 284         if is_explicit is True:
 285             age_limit = 18
 286         elif is_explicit is False:
 287             age_limit = 0
 288         else:
 289             age_limit = None
 290
 291         duration = video_info.get('duration')
 292
 293         return {
 294             'id': video_id,
 295             'title': title,
 296             'formats': formats,
 297             'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
 298             'timestamp': timestamp,
 299             'uploader': uploader,
 300             'duration': duration,
 301             'view_count': view_count,
 302             'age_limit': age_limit,
 303         }