_ Git - youtube-dl/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': '2dbc7e9fd4f1c60436c9aa73a5406193',
  30         'info_dict': {
  31             'id': 'Pt1kc_FniKM',
  32             'ext': 'mp4',
  33             'title': 'Hurts - Somebody to Die For',
  34             'description': 'md5:13e925b89af6b01c7e417332bd23c4bf',
  35             'uploader_id': 'HurtsVEVO',
  36             'uploader': 'HurtsVEVO',
  37             'upload_date': '20130624',
  38             'duration': 230,
  39         },
  40         'add_ie': ['Youtube'],
  41     }, {
  42         'note': 'v3 SMIL format',
  43         'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
  44         'md5': '13d5204f520af905eeffa675040b8e76',
  45         'info_dict': {
  46             'id': 'ByGmQn1uxJw',
  47             'ext': 'mp4',
  48             'title': 'Cassadee Pope - I Wish I Could Break Your Heart',
  49             'description': 'md5:5e9721c92ef117a6f69d00e9b42ceba7',
  50             'uploader_id': 'CassadeeVEVO',
  51             'uploader': 'CassadeeVEVO',
  52             'upload_date': '20140219',
  53             'duration': 226,
  54             'age_limit': 0,
  55         },
  56         'add_ie': ['Youtube'],
  57     }, {
  58         'note': 'Age-limited video',
  59         'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
  60         'info_dict': {
  61             'id': '07FYdnEawAQ',
  62             'ext': 'mp4',
  63             'age_limit': 18,
  64             'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
  65             'description': 'md5:64249768eec3bc4276236606ea996373',
  66             'uploader_id': 'justintimberlakeVEVO',
  67             'uploader': 'justintimberlakeVEVO',
  68             'upload_date': '20130703',
  69             'duration': 419,
  70         },
  71         'params': {
  72             'skip_download': 'true',
  73         },
  74         'add_ie': ['Youtube'],
  75     }, {
  76         'note': 'No video_info',
  77         'url': 'http://www.vevo.com/watch/k-camp-1/Till-I-Die/USUV71503000',
  78         'md5': 'a8b84d1d1957cd01046441b701b270fb',
  79         'info_dict': {
  80             'id': 'Lad2jHtJCqY',
  81             'ext': 'mp4',
  82             'title': 'K Camp - Till I Die ft. T.I.',
  83             'description': 'md5:0694920ededdee4a14cfc39695cc8ec3',
  84             'uploader_id': 'KCampVEVO',
  85             'uploader': 'KCampVEVO',
  86             'upload_date': '20151207',
  87             'duration': 193,
  88         },
  89         'add_ie': ['Youtube'],
  90     }]
  91     _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com'
  92     _SOURCE_TYPES = {
  93         0: 'youtube',
  94         1: 'brightcove',
  95         2: 'http',
  96         3: 'hls_ios',
  97         4: 'hls',
  98         5: 'smil',  # http
  99         7: 'f4m_cc',
 100         8: 'f4m_ak',
 101         9: 'f4m_l3',
 102         10: 'ism',
 103         13: 'smil',  # rtmp
 104         18: 'dash',
 105     }
 106     _VERSIONS = {
 107         0: 'youtube',  # only in AuthenticateVideo videoVersions
 108         1: 'level3',
 109         2: 'akamai',
 110         3: 'level3',
 111         4: 'amazon',
 112     }
 113
 114     def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
 115         formats = []
 116         els = smil.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
 117         for el in els:
 118             src = el.attrib['src']
 119             m = re.match(r'''(?xi)
 120                 (?P<ext>[a-z0-9]+):
 121                 (?P<path>
 122                     [/a-z0-9]+     # The directory and main part of the URL
 123                     _(?P<tbr>[0-9]+)k
 124                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 125                     _(?P<vcodec>[a-z0-9]+)
 126                     _(?P<vbr>[0-9]+)
 127                     _(?P<acodec>[a-z0-9]+)
 128                     _(?P<abr>[0-9]+)
 129                     \.[a-z0-9]+  # File extension
 130                 )''', src)
 131             if not m:
 132                 continue
 133
 134             format_url = self._SMIL_BASE_URL + m.group('path')
 135             formats.append({
 136                 'url': format_url,
 137                 'format_id': 'smil_' + m.group('tbr'),
 138                 'vcodec': m.group('vcodec'),
 139                 'acodec': m.group('acodec'),
 140                 'tbr': int(m.group('tbr')),
 141                 'vbr': int(m.group('vbr')),
 142                 'abr': int(m.group('abr')),
 143                 'ext': m.group('ext'),
 144                 'width': int(m.group('width')),
 145                 'height': int(m.group('height')),
 146             })
 147         return formats
 148
 149     def _initialize_api(self, video_url, video_id):
 150         req = sanitized_Request(
 151             'http://www.vevo.com/auth', data=b'')
 152         webpage = self._download_webpage(
 153             req, None,
 154             note='Retrieving oauth token',
 155             errnote='Unable to retrieve oauth token')
 156
 157         if 'THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION' in webpage:
 158             raise ExtractorError('%s said: This page is currently unavailable in your region.' % self.IE_NAME, expected=True)
 159
 160         auth_info = self._parse_json(webpage, video_id)
 161         self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['access_token']
 162
 163     def _call_api(self, path, video_id, note, errnote, fatal=True):
 164         return self._download_json(self._api_url_template % path, video_id, note, errnote)
 165
 166     def _real_extract(self, url):
 167         video_id = self._match_id(url)
 168
 169         json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
 170         response = self._download_json(json_url, video_id, 'Downloading video info', 'Unable to download info')
 171         video_info = response.get('video') or {}
 172         video_versions = video_info.get('videoVersions')
 173         uploader = None
 174         timestamp = None
 175         view_count = None
 176         formats = []
 177
 178         if not video_info:
 179             ytid = response.get('errorInfo', {}).get('ytid')
 180             if ytid:
 181                 return self.url_result(ytid, 'Youtube', ytid)
 182
 183             if response.get('statusCode') != 909:
 184                 if 'statusMessage' in response:
 185                     raise ExtractorError('%s said: %s' % (
 186                         self.IE_NAME, response['statusMessage']), expected=True)
 187                 raise ExtractorError('Unable to extract videos')
 188
 189             if url.startswith('vevo:'):
 190                 raise ExtractorError(
 191                     'Please specify full Vevo URL for downloading', expected=True)
 192
 193             self._initialize_api(url, video_id)
 194             video_info = self._call_api(
 195                 'video/%s' % video_id, video_id, 'Downloading api video info',
 196                 'Failed to download video info')
 197
 198             ytid = video_info.get('youTubeId')
 199             if ytid:
 200                 return self.url_result(
 201                     ytid, 'Youtube', ytid)
 202
 203             video_versions = self._call_api(
 204                 'video/%s/streams' % video_id, video_id,
 205                 'Downloading video versions info',
 206                 'Failed to download video versions info')
 207
 208             timestamp = parse_iso8601(video_info.get('releaseDate'))
 209             artists = video_info.get('artists')
 210             if artists:
 211                 uploader = artists[0]['name']
 212             view_count = int_or_none(video_info.get('views', {}).get('total'))
 213
 214             for video_version in video_versions:
 215                 version = self._VERSIONS.get(video_version['version'])
 216                 version_url = video_version.get('url')
 217                 if not version_url:
 218                         continue
 219
 220                 if '.mpd' in version_url or '.ism' in version_url:
 221                     continue
 222                 elif '.m3u8' in version_url:
 223                     formats.extend(self._extract_m3u8_formats(
 224                         version_url, video_id, 'mp4', 'm3u8_native',
 225                         m3u8_id='hls-%s' % version,
 226                         note='Downloading %s m3u8 information' % version,
 227                         errnote='Failed to download %s m3u8 information' % version,
 228                         fatal=False))
 229                 else:
 230                     m = re.search(r'''(?xi)
 231                         _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 232                         _(?P<vcodec>[a-z0-9]+)
 233                         _(?P<vbr>[0-9]+)
 234                         _(?P<acodec>[a-z0-9]+)
 235                         _(?P<abr>[0-9]+)
 236                         \.(?P<ext>[a-z0-9]+)''', version_url)
 237                     if not m:
 238                         continue
 239
 240                     formats.append({
 241                         'url': version_url,
 242                         'format_id': 'http-%s-%s' % (version, video_version['quality']),
 243                         'vcodec': m.group('vcodec'),
 244                         'acodec': m.group('acodec'),
 245                         'vbr': int(m.group('vbr')),
 246                         'abr': int(m.group('abr')),
 247                         'ext': m.group('ext'),
 248                         'width': int(m.group('width')),
 249                         'height': int(m.group('height')),
 250                     })
 251         else:
 252             timestamp = int_or_none(self._search_regex(
 253                 r'/Date\((\d+)\)/',
 254                 video_info['releaseDate'], 'release date', fatal=False),
 255                 scale=1000)
 256             artists = video_info.get('mainArtists')
 257             if artists:
 258                 uploader = artists[0]['artistName']
 259
 260             smil_parsed = False
 261             for video_version in video_info['videoVersions']:
 262                 version = self._VERSIONS.get(video_version['version'])
 263                 if version == 'youtube':
 264                     return self.url_result(
 265                         video_version['id'], 'Youtube', video_version['id'])
 266                 else:
 267                     source_type = self._SOURCE_TYPES.get(video_version['sourceType'])
 268                     renditions = compat_etree_fromstring(video_version['data'])
 269                     if source_type == 'http':
 270                         for rend in renditions.findall('rendition'):
 271                             attr = rend.attrib
 272                             formats.append({
 273                                 'url': attr['url'],
 274                                 'format_id': 'http-%s-%s' % (version, attr['name']),
 275                                 'height': int_or_none(attr.get('frameheight')),
 276                                 'width': int_or_none(attr.get('frameWidth')),
 277                                 'tbr': int_or_none(attr.get('totalBitrate')),
 278                                 'vbr': int_or_none(attr.get('videoBitrate')),
 279                                 'abr': int_or_none(attr.get('audioBitrate')),
 280                                 'vcodec': attr.get('videoCodec'),
 281                                 'acodec': attr.get('audioCodec'),
 282                             })
 283                     elif source_type == 'hls':
 284                         formats.extend(self._extract_m3u8_formats(
 285                             renditions.find('rendition').attrib['url'], video_id,
 286                             'mp4', 'm3u8_native', m3u8_id='hls-%s' % version,
 287                             note='Downloading %s m3u8 information' % version,
 288                             errnote='Failed to download %s m3u8 information' % version,
 289                             fatal=False))
 290                     elif source_type == 'smil' and not smil_parsed:
 291                         formats.extend(self._extract_smil_formats(
 292                             renditions.find('rendition').attrib['url'], video_id, False))
 293                         smil_parsed = True
 294         self._sort_formats(formats)
 295
 296         title = video_info['title']
 297
 298         is_explicit = video_info.get('isExplicit')
 299         if is_explicit is True:
 300             age_limit = 18
 301         elif is_explicit is False:
 302             age_limit = 0
 303         else:
 304             age_limit = None
 305
 306         duration = video_info.get('duration')
 307
 308         return {
 309             'id': video_id,
 310             'title': title,
 311             'formats': formats,
 312             'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
 313             'timestamp': timestamp,
 314             'uploader': uploader,
 315             'duration': duration,
 316             'view_count': view_count,
 317             'age_limit': age_limit,
 318         }