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