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