_ Git - youtube-dl/blob - youtube_dl/extractor/mixcloud.py

   1 from __future__ import unicode_literals
   2
   3 import functools
   4 import itertools
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_b64decode,
  10     compat_chr,
  11     compat_ord,
  12     compat_str,
  13     compat_urllib_parse_unquote,
  14     compat_urlparse,
  15     compat_zip
  16 )
  17 from ..utils import (
  18     clean_html,
  19     ExtractorError,
  20     int_or_none,
  21     OnDemandPagedList,
  22     str_to_int,
  23     try_get,
  24     urljoin,
  25 )
  26
  27
  28 class MixcloudIE(InfoExtractor):
  29     _VALID_URL = r'https?://(?:(?:www|beta|m)\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)'
  30     IE_NAME = 'mixcloud'
  31
  32     _TESTS = [{
  33         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
  34         'info_dict': {
  35             'id': 'dholbach-cryptkeeper',
  36             'ext': 'm4a',
  37             'title': 'Cryptkeeper',
  38             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
  39             'uploader': 'Daniel Holbach',
  40             'uploader_id': 'dholbach',
  41             'thumbnail': r're:https?://.*\.jpg',
  42             'view_count': int,
  43         },
  44     }, {
  45         'url': 'http://www.mixcloud.com/gillespeterson/caribou-7-inch-vinyl-mix-chat/',
  46         'info_dict': {
  47             'id': 'gillespeterson-caribou-7-inch-vinyl-mix-chat',
  48             'ext': 'mp3',
  49             'title': 'Caribou 7 inch Vinyl Mix & Chat',
  50             'description': 'md5:2b8aec6adce69f9d41724647c65875e8',
  51             'uploader': 'Gilles Peterson Worldwide',
  52             'uploader_id': 'gillespeterson',
  53             'thumbnail': 're:https?://.*',
  54             'view_count': int,
  55         },
  56     }, {
  57         'url': 'https://beta.mixcloud.com/RedLightRadio/nosedrip-15-red-light-radio-01-18-2016/',
  58         'only_matching': True,
  59     }]
  60
  61     @staticmethod
  62     def _decrypt_xor_cipher(key, ciphertext):
  63         """Encrypt/Decrypt XOR cipher. Both ways are possible because it's XOR."""
  64         return ''.join([
  65             compat_chr(compat_ord(ch) ^ compat_ord(k))
  66             for ch, k in compat_zip(ciphertext, itertools.cycle(key))])
  67
  68     def _real_extract(self, url):
  69         mobj = re.match(self._VALID_URL, url)
  70         uploader = mobj.group(1)
  71         cloudcast_name = mobj.group(2)
  72         track_id = compat_urllib_parse_unquote('-'.join((uploader, cloudcast_name)))
  73
  74         webpage = self._download_webpage(url, track_id)
  75
  76         # Legacy path
  77         encrypted_play_info = self._search_regex(
  78             r'm-play-info="([^"]+)"', webpage, 'play info', default=None)
  79
  80         if encrypted_play_info is not None:
  81             # Decode
  82             encrypted_play_info = compat_b64decode(encrypted_play_info)
  83         else:
  84             # New path
  85             full_info_json = self._parse_json(self._html_search_regex(
  86                 r'<script id="relay-data" type="text/x-mixcloud">([^<]+)</script>',
  87                 webpage, 'play info'), 'play info')
  88             for item in full_info_json:
  89                 item_data = try_get(item, [
  90                     lambda x: x['cloudcast']['data']['cloudcastLookup'],
  91                     lambda x: x['cloudcastLookup']['data']['cloudcastLookup'],
  92                 ], dict)
  93                 if try_get(item_data, lambda x: x['streamInfo']['url']):
  94                     info_json = item_data
  95                     break
  96             else:
  97                 raise ExtractorError('Failed to extract matching stream info')
  98
  99         message = self._html_search_regex(
 100             r'(?s)<div[^>]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)',
 101             webpage, 'error message', default=None)
 102
 103         js_url = self._search_regex(
 104             r'<script[^>]+\bsrc=["\"](https://(?:www\.)?mixcloud\.com/media/(?:js2/www_js_4|js/www)\.[^>]+\.js)',
 105             webpage, 'js url')
 106         js = self._download_webpage(js_url, track_id, 'Downloading JS')
 107         # Known plaintext attack
 108         if encrypted_play_info:
 109             kps = ['{"stream_url":']
 110             kpa_target = encrypted_play_info
 111         else:
 112             kps = ['https://', 'http://']
 113             kpa_target = compat_b64decode(info_json['streamInfo']['url'])
 114         for kp in kps:
 115             partial_key = self._decrypt_xor_cipher(kpa_target, kp)
 116             for quote in ["'", '"']:
 117                 key = self._search_regex(
 118                     r'{0}({1}[^{0}]*){0}'.format(quote, re.escape(partial_key)),
 119                     js, 'encryption key', default=None)
 120                 if key is not None:
 121                     break
 122             else:
 123                 continue
 124             break
 125         else:
 126             raise ExtractorError('Failed to extract encryption key')
 127
 128         if encrypted_play_info is not None:
 129             play_info = self._parse_json(self._decrypt_xor_cipher(key, encrypted_play_info), 'play info')
 130             if message and 'stream_url' not in play_info:
 131                 raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
 132             song_url = play_info['stream_url']
 133             formats = [{
 134                 'format_id': 'normal',
 135                 'url': song_url
 136             }]
 137
 138             title = self._html_search_regex(r'm-title="([^"]+)"', webpage, 'title')
 139             thumbnail = self._proto_relative_url(self._html_search_regex(
 140                 r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail', fatal=False))
 141             uploader = self._html_search_regex(
 142                 r'm-owner-name="([^"]+)"', webpage, 'uploader', fatal=False)
 143             uploader_id = self._search_regex(
 144                 r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
 145             description = self._og_search_description(webpage)
 146             view_count = str_to_int(self._search_regex(
 147                 [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
 148                  r'/listeners/?">([0-9,.]+)</a>',
 149                  r'(?:m|data)-tooltip=["\']([\d,.]+) plays'],
 150                 webpage, 'play count', default=None))
 151
 152         else:
 153             title = info_json['name']
 154             thumbnail = urljoin(
 155                 'https://thumbnailer.mixcloud.com/unsafe/600x600/',
 156                 try_get(info_json, lambda x: x['picture']['urlRoot'], compat_str))
 157             uploader = try_get(info_json, lambda x: x['owner']['displayName'])
 158             uploader_id = try_get(info_json, lambda x: x['owner']['username'])
 159             description = try_get(info_json, lambda x: x['description'])
 160             view_count = int_or_none(try_get(info_json, lambda x: x['plays']))
 161
 162             stream_info = info_json['streamInfo']
 163             formats = []
 164
 165             def decrypt_url(f_url):
 166                 for k in (key, 'IFYOUWANTTHEARTISTSTOGETPAIDDONOTDOWNLOADFROMMIXCLOUD'):
 167                     decrypted_url = self._decrypt_xor_cipher(k, f_url)
 168                     if re.search(r'^https?://[0-9A-Za-z.]+/[0-9A-Za-z/.?=&_-]+$', decrypted_url):
 169                         return decrypted_url
 170
 171             for url_key in ('url', 'hlsUrl', 'dashUrl'):
 172                 format_url = stream_info.get(url_key)
 173                 if not format_url:
 174                     continue
 175                 decrypted = decrypt_url(compat_b64decode(format_url))
 176                 if not decrypted:
 177                     continue
 178                 if url_key == 'hlsUrl':
 179                     formats.extend(self._extract_m3u8_formats(
 180                         decrypted, track_id, 'mp4', entry_protocol='m3u8_native',
 181                         m3u8_id='hls', fatal=False))
 182                 elif url_key == 'dashUrl':
 183                     formats.extend(self._extract_mpd_formats(
 184                         decrypted, track_id, mpd_id='dash', fatal=False))
 185                 else:
 186                     formats.append({
 187                         'format_id': 'http',
 188                         'url': decrypted,
 189                         'downloader_options': {
 190                             # Mixcloud starts throttling at >~5M
 191                             'http_chunk_size': 5242880,
 192                         },
 193                     })
 194             self._sort_formats(formats)
 195
 196         return {
 197             'id': track_id,
 198             'title': title,
 199             'formats': formats,
 200             'description': description,
 201             'thumbnail': thumbnail,
 202             'uploader': uploader,
 203             'uploader_id': uploader_id,
 204             'view_count': view_count,
 205         }
 206
 207
 208 class MixcloudPlaylistBaseIE(InfoExtractor):
 209     _PAGE_SIZE = 24
 210
 211     def _find_urls_in_page(self, page):
 212         for url in re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', page):
 213             yield self.url_result(
 214                 compat_urlparse.urljoin('https://www.mixcloud.com', clean_html(url)),
 215                 MixcloudIE.ie_key())
 216
 217     def _fetch_tracks_page(self, path, video_id, page_name, current_page, real_page_number=None):
 218         real_page_number = real_page_number or current_page + 1
 219         return self._download_webpage(
 220             'https://www.mixcloud.com/%s/' % path, video_id,
 221             note='Download %s (page %d)' % (page_name, current_page + 1),
 222             errnote='Unable to download %s' % page_name,
 223             query={'page': real_page_number, 'list': 'main', '_ajax': '1'},
 224             headers={'X-Requested-With': 'XMLHttpRequest'})
 225
 226     def _tracks_page_func(self, page, video_id, page_name, current_page):
 227         resp = self._fetch_tracks_page(page, video_id, page_name, current_page)
 228
 229         for item in self._find_urls_in_page(resp):
 230             yield item
 231
 232     def _get_user_description(self, page_content):
 233         return self._html_search_regex(
 234             r'<div[^>]+class="profile-bio"[^>]*>(.+?)</div>',
 235             page_content, 'user description', fatal=False)
 236
 237
 238 class MixcloudUserIE(MixcloudPlaylistBaseIE):
 239     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
 240     IE_NAME = 'mixcloud:user'
 241
 242     _TESTS = [{
 243         'url': 'http://www.mixcloud.com/dholbach/',
 244         'info_dict': {
 245             'id': 'dholbach_uploads',
 246             'title': 'Daniel Holbach (uploads)',
 247             'description': 'md5:def36060ac8747b3aabca54924897e47',
 248         },
 249         'playlist_mincount': 11,
 250     }, {
 251         'url': 'http://www.mixcloud.com/dholbach/uploads/',
 252         'info_dict': {
 253             'id': 'dholbach_uploads',
 254             'title': 'Daniel Holbach (uploads)',
 255             'description': 'md5:def36060ac8747b3aabca54924897e47',
 256         },
 257         'playlist_mincount': 11,
 258     }, {
 259         'url': 'http://www.mixcloud.com/dholbach/favorites/',
 260         'info_dict': {
 261             'id': 'dholbach_favorites',
 262             'title': 'Daniel Holbach (favorites)',
 263             'description': 'md5:def36060ac8747b3aabca54924897e47',
 264         },
 265         'params': {
 266             'playlist_items': '1-100',
 267         },
 268         'playlist_mincount': 100,
 269     }, {
 270         'url': 'http://www.mixcloud.com/dholbach/listens/',
 271         'info_dict': {
 272             'id': 'dholbach_listens',
 273             'title': 'Daniel Holbach (listens)',
 274             'description': 'md5:def36060ac8747b3aabca54924897e47',
 275         },
 276         'params': {
 277             'playlist_items': '1-100',
 278         },
 279         'playlist_mincount': 100,
 280     }]
 281
 282     def _real_extract(self, url):
 283         mobj = re.match(self._VALID_URL, url)
 284         user_id = mobj.group('user')
 285         list_type = mobj.group('type')
 286
 287         # if only a profile URL was supplied, default to download all uploads
 288         if list_type is None:
 289             list_type = 'uploads'
 290
 291         video_id = '%s_%s' % (user_id, list_type)
 292
 293         profile = self._download_webpage(
 294             'https://www.mixcloud.com/%s/' % user_id, video_id,
 295             note='Downloading user profile',
 296             errnote='Unable to download user profile')
 297
 298         username = self._og_search_title(profile)
 299         description = self._get_user_description(profile)
 300
 301         entries = OnDemandPagedList(
 302             functools.partial(
 303                 self._tracks_page_func,
 304                 '%s/%s' % (user_id, list_type), video_id, 'list of %s' % list_type),
 305             self._PAGE_SIZE)
 306
 307         return self.playlist_result(
 308             entries, video_id, '%s (%s)' % (username, list_type), description)
 309
 310
 311 class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
 312     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
 313     IE_NAME = 'mixcloud:playlist'
 314
 315     _TESTS = [{
 316         'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/',
 317         'info_dict': {
 318             'id': 'RedBullThre3style_tokyo-finalists-2015',
 319             'title': 'National Champions 2015',
 320             'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3',
 321         },
 322         'playlist_mincount': 16,
 323     }, {
 324         'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
 325         'only_matching': True,
 326     }]
 327
 328     def _real_extract(self, url):
 329         mobj = re.match(self._VALID_URL, url)
 330         user_id = mobj.group('user')
 331         playlist_id = mobj.group('playlist')
 332         video_id = '%s_%s' % (user_id, playlist_id)
 333
 334         webpage = self._download_webpage(
 335             url, user_id,
 336             note='Downloading playlist page',
 337             errnote='Unable to download playlist page')
 338
 339         title = self._html_search_regex(
 340             r'<a[^>]+class="parent active"[^>]*><b>\d+</b><span[^>]*>([^<]+)',
 341             webpage, 'playlist title',
 342             default=None) or self._og_search_title(webpage, fatal=False)
 343         description = self._get_user_description(webpage)
 344
 345         entries = OnDemandPagedList(
 346             functools.partial(
 347                 self._tracks_page_func,
 348                 '%s/playlists/%s' % (user_id, playlist_id), video_id, 'tracklist'),
 349             self._PAGE_SIZE)
 350
 351         return self.playlist_result(entries, video_id, title, description)
 352
 353
 354 class MixcloudStreamIE(MixcloudPlaylistBaseIE):
 355     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/stream/?$'
 356     IE_NAME = 'mixcloud:stream'
 357
 358     _TEST = {
 359         'url': 'https://www.mixcloud.com/FirstEar/stream/',
 360         'info_dict': {
 361             'id': 'FirstEar',
 362             'title': 'First Ear',
 363             'description': 'Curators of good music\nfirstearmusic.com',
 364         },
 365         'playlist_mincount': 192,
 366     }
 367
 368     def _real_extract(self, url):
 369         user_id = self._match_id(url)
 370
 371         webpage = self._download_webpage(url, user_id)
 372
 373         entries = []
 374         prev_page_url = None
 375
 376         def _handle_page(page):
 377             entries.extend(self._find_urls_in_page(page))
 378             return self._search_regex(
 379                 r'm-next-page-url="([^"]+)"', page,
 380                 'next page URL', default=None)
 381
 382         next_page_url = _handle_page(webpage)
 383
 384         for idx in itertools.count(0):
 385             if not next_page_url or prev_page_url == next_page_url:
 386                 break
 387
 388             prev_page_url = next_page_url
 389             current_page = int(self._search_regex(
 390                 r'\?page=(\d+)', next_page_url, 'next page number'))
 391
 392             next_page_url = _handle_page(self._fetch_tracks_page(
 393                 '%s/stream' % user_id, user_id, 'stream', idx,
 394                 real_page_number=current_page))
 395
 396         username = self._og_search_title(webpage)
 397         description = self._get_user_description(webpage)
 398
 399         return self.playlist_result(entries, user_id, username, description)