_ Git - youtube-dl/blob - 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 = self._search_regex(
  95                         r'player\s*:\s*{.*?\bvalue\s*:\s*(["\'])(?P<key>(?:(?!\1).)+)\1',
  96                         js, 'key', default=None, group='key')
  97                     if key and isinstance(key, compat_str):
  98                         self._keys.insert(0, key)
  99                         self._current_key = key
 100
 101         message = self._html_search_regex(
 102             r'(?s)<div[^>]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)',
 103             webpage, 'error message', default=None)
 104
 105         encrypted_play_info = self._search_regex(
 106             r'm-play-info="([^"]+)"', webpage, 'play info')
 107
 108         play_info = self._decrypt_play_info(encrypted_play_info, track_id)
 109
 110         if message and 'stream_url' not in play_info:
 111             raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
 112
 113         song_url = play_info['stream_url']
 114
 115         title = self._html_search_regex(r'm-title="([^"]+)"', webpage, 'title')
 116         thumbnail = self._proto_relative_url(self._html_search_regex(
 117             r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail', fatal=False))
 118         uploader = self._html_search_regex(
 119             r'm-owner-name="([^"]+)"', webpage, 'uploader', fatal=False)
 120         uploader_id = self._search_regex(
 121             r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
 122         description = self._og_search_description(webpage)
 123         view_count = str_to_int(self._search_regex(
 124             [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
 125              r'/listeners/?">([0-9,.]+)</a>',
 126              r'(?:m|data)-tooltip=["\']([\d,.]+) plays'],
 127             webpage, 'play count', default=None))
 128
 129         return {
 130             'id': track_id,
 131             'title': title,
 132             'url': song_url,
 133             'description': description,
 134             'thumbnail': thumbnail,
 135             'uploader': uploader,
 136             'uploader_id': uploader_id,
 137             'view_count': view_count,
 138         }
 139
 140
 141 class MixcloudPlaylistBaseIE(InfoExtractor):
 142     _PAGE_SIZE = 24
 143
 144     def _find_urls_in_page(self, page):
 145         for url in re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', page):
 146             yield self.url_result(
 147                 compat_urlparse.urljoin('https://www.mixcloud.com', clean_html(url)),
 148                 MixcloudIE.ie_key())
 149
 150     def _fetch_tracks_page(self, path, video_id, page_name, current_page, real_page_number=None):
 151         real_page_number = real_page_number or current_page + 1
 152         return self._download_webpage(
 153             'https://www.mixcloud.com/%s/' % path, video_id,
 154             note='Download %s (page %d)' % (page_name, current_page + 1),
 155             errnote='Unable to download %s' % page_name,
 156             query={'page': real_page_number, 'list': 'main', '_ajax': '1'},
 157             headers={'X-Requested-With': 'XMLHttpRequest'})
 158
 159     def _tracks_page_func(self, page, video_id, page_name, current_page):
 160         resp = self._fetch_tracks_page(page, video_id, page_name, current_page)
 161
 162         for item in self._find_urls_in_page(resp):
 163             yield item
 164
 165     def _get_user_description(self, page_content):
 166         return self._html_search_regex(
 167             r'<div[^>]+class="profile-bio"[^>]*>(.+?)</div>',
 168             page_content, 'user description', fatal=False)
 169
 170
 171 class MixcloudUserIE(MixcloudPlaylistBaseIE):
 172     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
 173     IE_NAME = 'mixcloud:user'
 174
 175     _TESTS = [{
 176         'url': 'http://www.mixcloud.com/dholbach/',
 177         'info_dict': {
 178             'id': 'dholbach_uploads',
 179             'title': 'Daniel Holbach (uploads)',
 180             'description': 'md5:def36060ac8747b3aabca54924897e47',
 181         },
 182         'playlist_mincount': 11,
 183     }, {
 184         'url': 'http://www.mixcloud.com/dholbach/uploads/',
 185         'info_dict': {
 186             'id': 'dholbach_uploads',
 187             'title': 'Daniel Holbach (uploads)',
 188             'description': 'md5:def36060ac8747b3aabca54924897e47',
 189         },
 190         'playlist_mincount': 11,
 191     }, {
 192         'url': 'http://www.mixcloud.com/dholbach/favorites/',
 193         'info_dict': {
 194             'id': 'dholbach_favorites',
 195             'title': 'Daniel Holbach (favorites)',
 196             'description': 'md5:def36060ac8747b3aabca54924897e47',
 197         },
 198         'params': {
 199             'playlist_items': '1-100',
 200         },
 201         'playlist_mincount': 100,
 202     }, {
 203         'url': 'http://www.mixcloud.com/dholbach/listens/',
 204         'info_dict': {
 205             'id': 'dholbach_listens',
 206             'title': 'Daniel Holbach (listens)',
 207             'description': 'md5:def36060ac8747b3aabca54924897e47',
 208         },
 209         'params': {
 210             'playlist_items': '1-100',
 211         },
 212         'playlist_mincount': 100,
 213     }]
 214
 215     def _real_extract(self, url):
 216         mobj = re.match(self._VALID_URL, url)
 217         user_id = mobj.group('user')
 218         list_type = mobj.group('type')
 219
 220         # if only a profile URL was supplied, default to download all uploads
 221         if list_type is None:
 222             list_type = 'uploads'
 223
 224         video_id = '%s_%s' % (user_id, list_type)
 225
 226         profile = self._download_webpage(
 227             'https://www.mixcloud.com/%s/' % user_id, video_id,
 228             note='Downloading user profile',
 229             errnote='Unable to download user profile')
 230
 231         username = self._og_search_title(profile)
 232         description = self._get_user_description(profile)
 233
 234         entries = OnDemandPagedList(
 235             functools.partial(
 236                 self._tracks_page_func,
 237                 '%s/%s' % (user_id, list_type), video_id, 'list of %s' % list_type),
 238             self._PAGE_SIZE, use_cache=True)
 239
 240         return self.playlist_result(
 241             entries, video_id, '%s (%s)' % (username, list_type), description)
 242
 243
 244 class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
 245     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
 246     IE_NAME = 'mixcloud:playlist'
 247
 248     _TESTS = [{
 249         'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/',
 250         'info_dict': {
 251             'id': 'RedBullThre3style_tokyo-finalists-2015',
 252             'title': 'National Champions 2015',
 253             'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3',
 254         },
 255         'playlist_mincount': 16,
 256     }, {
 257         'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
 258         'only_matching': True,
 259     }]
 260
 261     def _real_extract(self, url):
 262         mobj = re.match(self._VALID_URL, url)
 263         user_id = mobj.group('user')
 264         playlist_id = mobj.group('playlist')
 265         video_id = '%s_%s' % (user_id, playlist_id)
 266
 267         webpage = self._download_webpage(
 268             url, user_id,
 269             note='Downloading playlist page',
 270             errnote='Unable to download playlist page')
 271
 272         title = self._html_search_regex(
 273             r'<a[^>]+class="parent active"[^>]*><b>\d+</b><span[^>]*>([^<]+)',
 274             webpage, 'playlist title',
 275             default=None) or self._og_search_title(webpage, fatal=False)
 276         description = self._get_user_description(webpage)
 277
 278         entries = OnDemandPagedList(
 279             functools.partial(
 280                 self._tracks_page_func,
 281                 '%s/playlists/%s' % (user_id, playlist_id), video_id, 'tracklist'),
 282             self._PAGE_SIZE)
 283
 284         return self.playlist_result(entries, video_id, title, description)
 285
 286
 287 class MixcloudStreamIE(MixcloudPlaylistBaseIE):
 288     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/stream/?$'
 289     IE_NAME = 'mixcloud:stream'
 290
 291     _TEST = {
 292         'url': 'https://www.mixcloud.com/FirstEar/stream/',
 293         'info_dict': {
 294             'id': 'FirstEar',
 295             'title': 'First Ear',
 296             'description': 'Curators of good music\nfirstearmusic.com',
 297         },
 298         'playlist_mincount': 192,
 299     }
 300
 301     def _real_extract(self, url):
 302         user_id = self._match_id(url)
 303
 304         webpage = self._download_webpage(url, user_id)
 305
 306         entries = []
 307         prev_page_url = None
 308
 309         def _handle_page(page):
 310             entries.extend(self._find_urls_in_page(page))
 311             return self._search_regex(
 312                 r'm-next-page-url="([^"]+)"', page,
 313                 'next page URL', default=None)
 314
 315         next_page_url = _handle_page(webpage)
 316
 317         for idx in itertools.count(0):
 318             if not next_page_url or prev_page_url == next_page_url:
 319                 break
 320
 321             prev_page_url = next_page_url
 322             current_page = int(self._search_regex(
 323                 r'\?page=(\d+)', next_page_url, 'next page number'))
 324
 325             next_page_url = _handle_page(self._fetch_tracks_page(
 326                 '%s/stream' % user_id, user_id, 'stream', idx,
 327                 real_page_number=current_page))
 328
 329         username = self._og_search_title(webpage)
 330         description = self._get_user_description(webpage)
 331
 332         return self.playlist_result(entries, user_id, username, description)