youtube_dl/extractor/mixcloud.py

   1 from __future__ import unicode_literals
   2
   3 import base64
   4 import functools
   5 import itertools
   6 import re
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_chr,
  11     compat_ord,
  12     compat_str,
  13     compat_urllib_parse_unquote,
  14     compat_urlparse,
  15 )
  16 from ..utils import (
  17     clean_html,
  18     ExtractorError,
  19     OnDemandPagedList,
  20     str_to_int,
  21 )
  22
  23
  24 class MixcloudIE(InfoExtractor):
  25     _VALID_URL = r'https?://(?:(?:www|beta|m)\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)'
  26     IE_NAME = 'mixcloud'
  27
  28     _TESTS = [{
  29         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
  30         'info_dict': {
  31             'id': 'dholbach-cryptkeeper',
  32             'ext': 'm4a',
  33             'title': 'Cryptkeeper',
  34             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
  35             'uploader': 'Daniel Holbach',
  36             'uploader_id': 'dholbach',
  37             'thumbnail': r're:https?://.*\.jpg',
  38             'view_count': int,
  39         },
  40     }, {
  41         'url': 'http://www.mixcloud.com/gillespeterson/caribou-7-inch-vinyl-mix-chat/',
  42         'info_dict': {
  43             'id': 'gillespeterson-caribou-7-inch-vinyl-mix-chat',
  44             'ext': 'mp3',
  45             'title': 'Caribou 7 inch Vinyl Mix & Chat',
  46             'description': 'md5:2b8aec6adce69f9d41724647c65875e8',
  47             'uploader': 'Gilles Peterson Worldwide',
  48             'uploader_id': 'gillespeterson',
  49             'thumbnail': 're:https?://.*',
  50             'view_count': int,
  51         },
  52     }, {
  53         'url': 'https://beta.mixcloud.com/RedLightRadio/nosedrip-15-red-light-radio-01-18-2016/',
  54         'only_matching': True,
  55     }]
  56
  57     _keys = [
  58         'return { requestAnimationFrame: function(callback) { callback(); }, innerHeight: 500 };',
  59         'pleasedontdownloadourmusictheartistswontgetpaid',
  60         'window.addEventListener = window.addEventListener || function() {};',
  61         '(function() { return new Date().toLocaleDateString(); })()'
  62     ]
  63     _current_key = None
  64
  65     # See https://www.mixcloud.com/media/js2/www_js_2.9e23256562c080482435196ca3975ab5.js
  66     def _decrypt_play_info(self, play_info, video_id):
  67         play_info = base64.b64decode(play_info.encode('ascii'))
  68         for num, key in enumerate(self._keys, start=1):
  69             try:
  70                 return self._parse_json(
  71                     ''.join([
  72                         compat_chr(compat_ord(ch) ^ compat_ord(key[idx % len(key)]))
  73                         for idx, ch in enumerate(play_info)]),
  74                     video_id)
  75             except ExtractorError:
  76                 if num == len(self._keys):
  77                     raise
  78
  79     def _real_extract(self, url):
  80         mobj = re.match(self._VALID_URL, url)
  81         uploader = mobj.group(1)
  82         cloudcast_name = mobj.group(2)
  83         track_id = compat_urllib_parse_unquote('-'.join((uploader, cloudcast_name)))
  84
  85         webpage = self._download_webpage(url, track_id)
  86
  87         if not self._current_key:
  88             js_url = self._search_regex(
  89                 r'<script[^>]+\bsrc=["\"](https://(?:www\.)?mixcloud\.com/media/js2/www_js_4\.[^>]+\.js)',
  90                 webpage, 'js url', default=None)
  91             if js_url:
  92                 js = self._download_webpage(js_url, track_id, fatal=False)
  93                 if js:
  94                     KEY_RE_TEMPLATE = r'player\s*:\s*{.*?\b%s\s*:\s*(["\'])(?P<key>(?:(?!\1).)+)\1'
  95                     for key_name in ('value', 'key_value'):
  96                         key = self._search_regex(
  97                             KEY_RE_TEMPLATE % key_name, js, 'key',
  98                             default=None, group='key')
  99                         if key and isinstance(key, compat_str):
 100                             self._keys.insert(0, key)
 101                             self._current_key = key
 102
 103         message = self._html_search_regex(
 104             r'(?s)<div[^>]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)',
 105             webpage, 'error message', default=None)
 106
 107         encrypted_play_info = self._search_regex(
 108             r'm-play-info="([^"]+)"', webpage, 'play info')
 109
 110         play_info = self._decrypt_play_info(encrypted_play_info, track_id)
 111
 112         if message and 'stream_url' not in play_info:
 113             raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
 114
 115         song_url = play_info['stream_url']
 116
 117         title = self._html_search_regex(r'm-title="([^"]+)"', webpage, 'title')
 118         thumbnail = self._proto_relative_url(self._html_search_regex(
 119             r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail', fatal=False))
 120         uploader = self._html_search_regex(
 121             r'm-owner-name="([^"]+)"', webpage, 'uploader', fatal=False)
 122         uploader_id = self._search_regex(
 123             r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
 124         description = self._og_search_description(webpage)
 125         view_count = str_to_int(self._search_regex(
 126             [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
 127              r'/listeners/?">([0-9,.]+)</a>',
 128              r'(?:m|data)-tooltip=["\']([\d,.]+) plays'],
 129             webpage, 'play count', default=None))
 130
 131         return {
 132             'id': track_id,
 133             'title': title,
 134             'url': song_url,
 135             'description': description,
 136             'thumbnail': thumbnail,
 137             'uploader': uploader,
 138             'uploader_id': uploader_id,
 139             'view_count': view_count,
 140         }
 141
 142
 143 class MixcloudPlaylistBaseIE(InfoExtractor):
 144     _PAGE_SIZE = 24
 145
 146     def _find_urls_in_page(self, page):
 147         for url in re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', page):
 148             yield self.url_result(
 149                 compat_urlparse.urljoin('https://www.mixcloud.com', clean_html(url)),
 150                 MixcloudIE.ie_key())
 151
 152     def _fetch_tracks_page(self, path, video_id, page_name, current_page, real_page_number=None):
 153         real_page_number = real_page_number or current_page + 1
 154         return self._download_webpage(
 155             'https://www.mixcloud.com/%s/' % path, video_id,
 156             note='Download %s (page %d)' % (page_name, current_page + 1),
 157             errnote='Unable to download %s' % page_name,
 158             query={'page': real_page_number, 'list': 'main', '_ajax': '1'},
 159             headers={'X-Requested-With': 'XMLHttpRequest'})
 160
 161     def _tracks_page_func(self, page, video_id, page_name, current_page):
 162         resp = self._fetch_tracks_page(page, video_id, page_name, current_page)
 163
 164         for item in self._find_urls_in_page(resp):
 165             yield item
 166
 167     def _get_user_description(self, page_content):
 168         return self._html_search_regex(
 169             r'<div[^>]+class="profile-bio"[^>]*>(.+?)</div>',
 170             page_content, 'user description', fatal=False)
 171
 172
 173 class MixcloudUserIE(MixcloudPlaylistBaseIE):
 174     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
 175     IE_NAME = 'mixcloud:user'
 176
 177     _TESTS = [{
 178         'url': 'http://www.mixcloud.com/dholbach/',
 179         'info_dict': {
 180             'id': 'dholbach_uploads',
 181             'title': 'Daniel Holbach (uploads)',
 182             'description': 'md5:def36060ac8747b3aabca54924897e47',
 183         },
 184         'playlist_mincount': 11,
 185     }, {
 186         'url': 'http://www.mixcloud.com/dholbach/uploads/',
 187         'info_dict': {
 188             'id': 'dholbach_uploads',
 189             'title': 'Daniel Holbach (uploads)',
 190             'description': 'md5:def36060ac8747b3aabca54924897e47',
 191         },
 192         'playlist_mincount': 11,
 193     }, {
 194         'url': 'http://www.mixcloud.com/dholbach/favorites/',
 195         'info_dict': {
 196             'id': 'dholbach_favorites',
 197             'title': 'Daniel Holbach (favorites)',
 198             'description': 'md5:def36060ac8747b3aabca54924897e47',
 199         },
 200         'params': {
 201             'playlist_items': '1-100',
 202         },
 203         'playlist_mincount': 100,
 204     }, {
 205         'url': 'http://www.mixcloud.com/dholbach/listens/',
 206         'info_dict': {
 207             'id': 'dholbach_listens',
 208             'title': 'Daniel Holbach (listens)',
 209             'description': 'md5:def36060ac8747b3aabca54924897e47',
 210         },
 211         'params': {
 212             'playlist_items': '1-100',
 213         },
 214         'playlist_mincount': 100,
 215     }]
 216
 217     def _real_extract(self, url):
 218         mobj = re.match(self._VALID_URL, url)
 219         user_id = mobj.group('user')
 220         list_type = mobj.group('type')
 221
 222         # if only a profile URL was supplied, default to download all uploads
 223         if list_type is None:
 224             list_type = 'uploads'
 225
 226         video_id = '%s_%s' % (user_id, list_type)
 227
 228         profile = self._download_webpage(
 229             'https://www.mixcloud.com/%s/' % user_id, video_id,
 230             note='Downloading user profile',
 231             errnote='Unable to download user profile')
 232
 233         username = self._og_search_title(profile)
 234         description = self._get_user_description(profile)
 235
 236         entries = OnDemandPagedList(
 237             functools.partial(
 238                 self._tracks_page_func,
 239                 '%s/%s' % (user_id, list_type), video_id, 'list of %s' % list_type),
 240             self._PAGE_SIZE, use_cache=True)
 241
 242         return self.playlist_result(
 243             entries, video_id, '%s (%s)' % (username, list_type), description)
 244
 245
 246 class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
 247     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
 248     IE_NAME = 'mixcloud:playlist'
 249
 250     _TESTS = [{
 251         'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/',
 252         'info_dict': {
 253             'id': 'RedBullThre3style_tokyo-finalists-2015',
 254             'title': 'National Champions 2015',
 255             'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3',
 256         },
 257         'playlist_mincount': 16,
 258     }, {
 259         'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
 260         'only_matching': True,
 261     }]
 262
 263     def _real_extract(self, url):
 264         mobj = re.match(self._VALID_URL, url)
 265         user_id = mobj.group('user')
 266         playlist_id = mobj.group('playlist')
 267         video_id = '%s_%s' % (user_id, playlist_id)
 268
 269         webpage = self._download_webpage(
 270             url, user_id,
 271             note='Downloading playlist page',
 272             errnote='Unable to download playlist page')
 273
 274         title = self._html_search_regex(
 275             r'<a[^>]+class="parent active"[^>]*><b>\d+</b><span[^>]*>([^<]+)',
 276             webpage, 'playlist title',
 277             default=None) or self._og_search_title(webpage, fatal=False)
 278         description = self._get_user_description(webpage)
 279
 280         entries = OnDemandPagedList(
 281             functools.partial(
 282                 self._tracks_page_func,
 283                 '%s/playlists/%s' % (user_id, playlist_id), video_id, 'tracklist'),
 284             self._PAGE_SIZE)
 285
 286         return self.playlist_result(entries, video_id, title, description)
 287
 288
 289 class MixcloudStreamIE(MixcloudPlaylistBaseIE):
 290     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/stream/?$'
 291     IE_NAME = 'mixcloud:stream'
 292
 293     _TEST = {
 294         'url': 'https://www.mixcloud.com/FirstEar/stream/',
 295         'info_dict': {
 296             'id': 'FirstEar',
 297             'title': 'First Ear',
 298             'description': 'Curators of good music\nfirstearmusic.com',
 299         },
 300         'playlist_mincount': 192,
 301     }
 302
 303     def _real_extract(self, url):
 304         user_id = self._match_id(url)
 305
 306         webpage = self._download_webpage(url, user_id)
 307
 308         entries = []
 309         prev_page_url = None
 310
 311         def _handle_page(page):
 312             entries.extend(self._find_urls_in_page(page))
 313             return self._search_regex(
 314                 r'm-next-page-url="([^"]+)"', page,
 315                 'next page URL', default=None)
 316
 317         next_page_url = _handle_page(webpage)
 318
 319         for idx in itertools.count(0):
 320             if not next_page_url or prev_page_url == next_page_url:
 321                 break
 322
 323             prev_page_url = next_page_url
 324             current_page = int(self._search_regex(
 325                 r'\?page=(\d+)', next_page_url, 'next page number'))
 326
 327             next_page_url = _handle_page(self._fetch_tracks_page(
 328                 '%s/stream' % user_id, user_id, 'stream', idx,
 329                 real_page_number=current_page))
 330
 331         username = self._og_search_title(webpage)
 332         description = self._get_user_description(webpage)
 333
 334         return self.playlist_result(entries, user_id, username, description)