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     @staticmethod
  58     def _decrypt_play_info(play_info):
  59         KEY = 'pleasedontdownloadourmusictheartistswontgetpaid'
  60
  61         play_info = base64.b64decode(play_info.encode('ascii'))
  62
  63         return ''.join([
  64             compat_chr(compat_ord(ch) ^ compat_ord(KEY[idx % len(KEY)]))
  65             for idx, ch in enumerate(play_info)])
  66
  67     def _real_extract(self, url):
  68         mobj = re.match(self._VALID_URL, url)
  69         uploader = mobj.group(1)
  70         cloudcast_name = mobj.group(2)
  71         track_id = compat_urllib_parse_unquote('-'.join((uploader, cloudcast_name)))
  72
  73         webpage = self._download_webpage(url, track_id)
  74
  75         message = self._html_search_regex(
  76             r'(?s)<div[^>]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)',
  77             webpage, 'error message', default=None)
  78
  79         encrypted_play_info = self._search_regex(
  80             r'm-play-info="([^"]+)"', webpage, 'play info')
  81         play_info = self._parse_json(
  82             self._decrypt_play_info(encrypted_play_info), track_id)
  83
  84         if message and 'stream_url' not in play_info:
  85             raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
  86
  87         song_url = play_info['stream_url']
  88
  89         title = self._html_search_regex(r'm-title="([^"]+)"', webpage, 'title')
  90         thumbnail = self._proto_relative_url(self._html_search_regex(
  91             r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail', fatal=False))
  92         uploader = self._html_search_regex(
  93             r'm-owner-name="([^"]+)"', webpage, 'uploader', fatal=False)
  94         uploader_id = self._search_regex(
  95             r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
  96         description = self._og_search_description(webpage)
  97         view_count = str_to_int(self._search_regex(
  98             [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
  99              r'/listeners/?">([0-9,.]+)</a>',
 100              r'm-tooltip=["\']([\d,.]+) plays'],
 101             webpage, 'play count', default=None))
 102
 103         return {
 104             'id': track_id,
 105             'title': title,
 106             'url': song_url,
 107             'description': description,
 108             'thumbnail': thumbnail,
 109             'uploader': uploader,
 110             'uploader_id': uploader_id,
 111             'view_count': view_count,
 112         }
 113
 114
 115 class MixcloudPlaylistBaseIE(InfoExtractor):
 116     _PAGE_SIZE = 24
 117
 118     def _find_urls_in_page(self, page):
 119         for url in re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', page):
 120             yield self.url_result(
 121                 compat_urlparse.urljoin('https://www.mixcloud.com', clean_html(url)),
 122                 MixcloudIE.ie_key())
 123
 124     def _fetch_tracks_page(self, path, video_id, page_name, current_page, real_page_number=None):
 125         real_page_number = real_page_number or current_page + 1
 126         return self._download_webpage(
 127             'https://www.mixcloud.com/%s/' % path, video_id,
 128             note='Download %s (page %d)' % (page_name, current_page + 1),
 129             errnote='Unable to download %s' % page_name,
 130             query={'page': real_page_number, 'list': 'main', '_ajax': '1'},
 131             headers={'X-Requested-With': 'XMLHttpRequest'})
 132
 133     def _tracks_page_func(self, page, video_id, page_name, current_page):
 134         resp = self._fetch_tracks_page(page, video_id, page_name, current_page)
 135
 136         for item in self._find_urls_in_page(resp):
 137             yield item
 138
 139     def _get_user_description(self, page_content):
 140         return self._html_search_regex(
 141             r'<div[^>]+class="description-text"[^>]*>(.+?)</div>',
 142             page_content, 'user description', fatal=False)
 143
 144
 145 class MixcloudUserIE(MixcloudPlaylistBaseIE):
 146     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
 147     IE_NAME = 'mixcloud:user'
 148
 149     _TESTS = [{
 150         'url': 'http://www.mixcloud.com/dholbach/',
 151         'info_dict': {
 152             'id': 'dholbach_uploads',
 153             'title': 'Daniel Holbach (uploads)',
 154             'description': 'md5:327af72d1efeb404a8216c27240d1370',
 155         },
 156         'playlist_mincount': 11,
 157     }, {
 158         'url': 'http://www.mixcloud.com/dholbach/uploads/',
 159         'info_dict': {
 160             'id': 'dholbach_uploads',
 161             'title': 'Daniel Holbach (uploads)',
 162             'description': 'md5:327af72d1efeb404a8216c27240d1370',
 163         },
 164         'playlist_mincount': 11,
 165     }, {
 166         'url': 'http://www.mixcloud.com/dholbach/favorites/',
 167         'info_dict': {
 168             'id': 'dholbach_favorites',
 169             'title': 'Daniel Holbach (favorites)',
 170             'description': 'md5:327af72d1efeb404a8216c27240d1370',
 171         },
 172         'params': {
 173             'playlist_items': '1-100',
 174         },
 175         'playlist_mincount': 100,
 176     }, {
 177         'url': 'http://www.mixcloud.com/dholbach/listens/',
 178         'info_dict': {
 179             'id': 'dholbach_listens',
 180             'title': 'Daniel Holbach (listens)',
 181             'description': 'md5:327af72d1efeb404a8216c27240d1370',
 182         },
 183         'params': {
 184             'playlist_items': '1-100',
 185         },
 186         'playlist_mincount': 100,
 187     }]
 188
 189     def _real_extract(self, url):
 190         mobj = re.match(self._VALID_URL, url)
 191         user_id = mobj.group('user')
 192         list_type = mobj.group('type')
 193
 194         # if only a profile URL was supplied, default to download all uploads
 195         if list_type is None:
 196             list_type = 'uploads'
 197
 198         video_id = '%s_%s' % (user_id, list_type)
 199
 200         profile = self._download_webpage(
 201             'https://www.mixcloud.com/%s/' % user_id, video_id,
 202             note='Downloading user profile',
 203             errnote='Unable to download user profile')
 204
 205         username = self._og_search_title(profile)
 206         description = self._get_user_description(profile)
 207
 208         entries = OnDemandPagedList(
 209             functools.partial(
 210                 self._tracks_page_func,
 211                 '%s/%s' % (user_id, list_type), video_id, 'list of %s' % list_type),
 212             self._PAGE_SIZE, use_cache=True)
 213
 214         return self.playlist_result(
 215             entries, video_id, '%s (%s)' % (username, list_type), description)
 216
 217
 218 class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
 219     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
 220     IE_NAME = 'mixcloud:playlist'
 221
 222     _TESTS = [{
 223         'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/',
 224         'info_dict': {
 225             'id': 'RedBullThre3style_tokyo-finalists-2015',
 226             'title': 'National Champions 2015',
 227             'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3',
 228         },
 229         'playlist_mincount': 16,
 230     }, {
 231         'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
 232         'info_dict': {
 233             'id': 'maxvibes_jazzcat-on-ness-radio',
 234             'title': 'Jazzcat on Ness Radio',
 235             'description': 'md5:7bbbf0d6359a0b8cda85224be0f8f263',
 236         },
 237         'playlist_mincount': 23
 238     }]
 239
 240     def _real_extract(self, url):
 241         mobj = re.match(self._VALID_URL, url)
 242         user_id = mobj.group('user')
 243         playlist_id = mobj.group('playlist')
 244         video_id = '%s_%s' % (user_id, playlist_id)
 245
 246         profile = self._download_webpage(
 247             url, user_id,
 248             note='Downloading playlist page',
 249             errnote='Unable to download playlist page')
 250
 251         description = self._get_user_description(profile)
 252         playlist_title = self._html_search_regex(
 253             r'<span[^>]+class="[^"]*list-playlist-title[^"]*"[^>]*>(.*?)</span>',
 254             profile, 'playlist title')
 255
 256         entries = OnDemandPagedList(
 257             functools.partial(
 258                 self._tracks_page_func,
 259                 '%s/playlists/%s' % (user_id, playlist_id), video_id, 'tracklist'),
 260             self._PAGE_SIZE)
 261
 262         return self.playlist_result(entries, video_id, playlist_title, description)
 263
 264
 265 class MixcloudStreamIE(MixcloudPlaylistBaseIE):
 266     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/stream/?$'
 267     IE_NAME = 'mixcloud:stream'
 268
 269     _TEST = {
 270         'url': 'https://www.mixcloud.com/FirstEar/stream/',
 271         'info_dict': {
 272             'id': 'FirstEar',
 273             'title': 'First Ear',
 274             'description': 'Curators of good music\nfirstearmusic.com',
 275         },
 276         'playlist_mincount': 192,
 277     }
 278
 279     def _real_extract(self, url):
 280         user_id = self._match_id(url)
 281
 282         webpage = self._download_webpage(url, user_id)
 283
 284         entries = []
 285         prev_page_url = None
 286
 287         def _handle_page(page):
 288             entries.extend(self._find_urls_in_page(page))
 289             return self._search_regex(
 290                 r'm-next-page-url="([^"]+)"', page,
 291                 'next page URL', default=None)
 292
 293         next_page_url = _handle_page(webpage)
 294
 295         for idx in itertools.count(0):
 296             if not next_page_url or prev_page_url == next_page_url:
 297                 break
 298
 299             prev_page_url = next_page_url
 300             current_page = int(self._search_regex(
 301                 r'\?page=(\d+)', next_page_url, 'next page number'))
 302
 303             next_page_url = _handle_page(self._fetch_tracks_page(
 304                 '%s/stream' % user_id, user_id, 'stream', idx,
 305                 real_page_number=current_page))
 306
 307         username = self._og_search_title(webpage)
 308         description = self._get_user_description(webpage)
 309
 310         return self.playlist_result(entries, user_id, username, description)