Raphaël G. Git Repositories - youtubedl/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(
  90                     item, lambda x: x['cloudcast']['data']['cloudcastLookup'],
  91                     dict)
  92                 if try_get(item_data, lambda x: x['streamInfo']['url']):
  93                     info_json = item_data
  94                     break
  95             else:
  96                 raise ExtractorError('Failed to extract matching stream info')
  97
  98         message = self._html_search_regex(
  99             r'(?s)<div[^>]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)',
 100             webpage, 'error message', default=None)
 101
 102         js_url = self._search_regex(
 103             r'<script[^>]+\bsrc=["\"](https://(?:www\.)?mixcloud\.com/media/(?:js2/www_js_4|js/www)\.[^>]+\.js)',
 104             webpage, 'js url')
 105         js = self._download_webpage(js_url, track_id, 'Downloading JS')
 106         # Known plaintext attack
 107         if encrypted_play_info:
 108             kps = ['{"stream_url":']
 109             kpa_target = encrypted_play_info
 110         else:
 111             kps = ['https://', 'http://']
 112             kpa_target = compat_b64decode(info_json['streamInfo']['url'])
 113         for kp in kps:
 114             partial_key = self._decrypt_xor_cipher(kpa_target, kp)
 115             for quote in ["'", '"']:
 116                 key = self._search_regex(
 117                     r'{0}({1}[^{0}]*){0}'.format(quote, re.escape(partial_key)),
 118                     js, 'encryption key', default=None)
 119                 if key is not None:
 120                     break
 121             else:
 122                 continue
 123             break
 124         else:
 125             raise ExtractorError('Failed to extract encryption key')
 126
 127         if encrypted_play_info is not None:
 128             play_info = self._parse_json(self._decrypt_xor_cipher(key, encrypted_play_info), 'play info')
 129             if message and 'stream_url' not in play_info:
 130                 raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
 131             song_url = play_info['stream_url']
 132             formats = [{
 133                 'format_id': 'normal',
 134                 'url': song_url
 135             }]
 136
 137             title = self._html_search_regex(r'm-title="([^"]+)"', webpage, 'title')
 138             thumbnail = self._proto_relative_url(self._html_search_regex(
 139                 r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail', fatal=False))
 140             uploader = self._html_search_regex(
 141                 r'm-owner-name="([^"]+)"', webpage, 'uploader', fatal=False)
 142             uploader_id = self._search_regex(
 143                 r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
 144             description = self._og_search_description(webpage)
 145             view_count = str_to_int(self._search_regex(
 146                 [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
 147                  r'/listeners/?">([0-9,.]+)</a>',
 148                  r'(?:m|data)-tooltip=["\']([\d,.]+) plays'],
 149                 webpage, 'play count', default=None))
 150
 151         else:
 152             title = info_json['name']
 153             thumbnail = urljoin(
 154                 'https://thumbnailer.mixcloud.com/unsafe/600x600/',
 155                 try_get(info_json, lambda x: x['picture']['urlRoot'], compat_str))
 156             uploader = try_get(info_json, lambda x: x['owner']['displayName'])
 157             uploader_id = try_get(info_json, lambda x: x['owner']['username'])
 158             description = try_get(info_json, lambda x: x['description'])
 159             view_count = int_or_none(try_get(info_json, lambda x: x['plays']))
 160
 161             stream_info = info_json['streamInfo']
 162             formats = []
 163
 164             for url_key in ('url', 'hlsUrl', 'dashUrl'):
 165                 format_url = stream_info.get(url_key)
 166                 if not format_url:
 167                     continue
 168                 decrypted = self._decrypt_xor_cipher(key, compat_b64decode(format_url))
 169                 if not decrypted:
 170                     continue
 171                 if url_key == 'hlsUrl':
 172                     formats.extend(self._extract_m3u8_formats(
 173                         decrypted, track_id, 'mp4', entry_protocol='m3u8_native',
 174                         m3u8_id='hls', fatal=False))
 175                 elif url_key == 'dashUrl':
 176                     formats.extend(self._extract_mpd_formats(
 177                         decrypted, track_id, mpd_id='dash', fatal=False))
 178                 else:
 179                     formats.append({
 180                         'format_id': 'http',
 181                         'url': decrypted,
 182                         'downloader_options': {
 183                             # Mixcloud starts throttling at >~5M
 184                             'http_chunk_size': 5242880,
 185                         },
 186                     })
 187             self._sort_formats(formats)
 188
 189         return {
 190             'id': track_id,
 191             'title': title,
 192             'formats': formats,
 193             'description': description,
 194             'thumbnail': thumbnail,
 195             'uploader': uploader,
 196             'uploader_id': uploader_id,
 197             'view_count': view_count,
 198         }
 199
 200
 201 class MixcloudPlaylistBaseIE(InfoExtractor):
 202     _PAGE_SIZE = 24
 203
 204     def _find_urls_in_page(self, page):
 205         for url in re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', page):
 206             yield self.url_result(
 207                 compat_urlparse.urljoin('https://www.mixcloud.com', clean_html(url)),
 208                 MixcloudIE.ie_key())
 209
 210     def _fetch_tracks_page(self, path, video_id, page_name, current_page, real_page_number=None):
 211         real_page_number = real_page_number or current_page + 1
 212         return self._download_webpage(
 213             'https://www.mixcloud.com/%s/' % path, video_id,
 214             note='Download %s (page %d)' % (page_name, current_page + 1),
 215             errnote='Unable to download %s' % page_name,
 216             query={'page': real_page_number, 'list': 'main', '_ajax': '1'},
 217             headers={'X-Requested-With': 'XMLHttpRequest'})
 218
 219     def _tracks_page_func(self, page, video_id, page_name, current_page):
 220         resp = self._fetch_tracks_page(page, video_id, page_name, current_page)
 221
 222         for item in self._find_urls_in_page(resp):
 223             yield item
 224
 225     def _get_user_description(self, page_content):
 226         return self._html_search_regex(
 227             r'<div[^>]+class="profile-bio"[^>]*>(.+?)</div>',
 228             page_content, 'user description', fatal=False)
 229
 230
 231 class MixcloudUserIE(MixcloudPlaylistBaseIE):
 232     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
 233     IE_NAME = 'mixcloud:user'
 234
 235     _TESTS = [{
 236         'url': 'http://www.mixcloud.com/dholbach/',
 237         'info_dict': {
 238             'id': 'dholbach_uploads',
 239             'title': 'Daniel Holbach (uploads)',
 240             'description': 'md5:def36060ac8747b3aabca54924897e47',
 241         },
 242         'playlist_mincount': 11,
 243     }, {
 244         'url': 'http://www.mixcloud.com/dholbach/uploads/',
 245         'info_dict': {
 246             'id': 'dholbach_uploads',
 247             'title': 'Daniel Holbach (uploads)',
 248             'description': 'md5:def36060ac8747b3aabca54924897e47',
 249         },
 250         'playlist_mincount': 11,
 251     }, {
 252         'url': 'http://www.mixcloud.com/dholbach/favorites/',
 253         'info_dict': {
 254             'id': 'dholbach_favorites',
 255             'title': 'Daniel Holbach (favorites)',
 256             'description': 'md5:def36060ac8747b3aabca54924897e47',
 257         },
 258         'params': {
 259             'playlist_items': '1-100',
 260         },
 261         'playlist_mincount': 100,
 262     }, {
 263         'url': 'http://www.mixcloud.com/dholbach/listens/',
 264         'info_dict': {
 265             'id': 'dholbach_listens',
 266             'title': 'Daniel Holbach (listens)',
 267             'description': 'md5:def36060ac8747b3aabca54924897e47',
 268         },
 269         'params': {
 270             'playlist_items': '1-100',
 271         },
 272         'playlist_mincount': 100,
 273     }]
 274
 275     def _real_extract(self, url):
 276         mobj = re.match(self._VALID_URL, url)
 277         user_id = mobj.group('user')
 278         list_type = mobj.group('type')
 279
 280         # if only a profile URL was supplied, default to download all uploads
 281         if list_type is None:
 282             list_type = 'uploads'
 283
 284         video_id = '%s_%s' % (user_id, list_type)
 285
 286         profile = self._download_webpage(
 287             'https://www.mixcloud.com/%s/' % user_id, video_id,
 288             note='Downloading user profile',
 289             errnote='Unable to download user profile')
 290
 291         username = self._og_search_title(profile)
 292         description = self._get_user_description(profile)
 293
 294         entries = OnDemandPagedList(
 295             functools.partial(
 296                 self._tracks_page_func,
 297                 '%s/%s' % (user_id, list_type), video_id, 'list of %s' % list_type),
 298             self._PAGE_SIZE)
 299
 300         return self.playlist_result(
 301             entries, video_id, '%s (%s)' % (username, list_type), description)
 302
 303
 304 class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
 305     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
 306     IE_NAME = 'mixcloud:playlist'
 307
 308     _TESTS = [{
 309         'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/',
 310         'info_dict': {
 311             'id': 'RedBullThre3style_tokyo-finalists-2015',
 312             'title': 'National Champions 2015',
 313             'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3',
 314         },
 315         'playlist_mincount': 16,
 316     }, {
 317         'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
 318         'only_matching': True,
 319     }]
 320
 321     def _real_extract(self, url):
 322         mobj = re.match(self._VALID_URL, url)
 323         user_id = mobj.group('user')
 324         playlist_id = mobj.group('playlist')
 325         video_id = '%s_%s' % (user_id, playlist_id)
 326
 327         webpage = self._download_webpage(
 328             url, user_id,
 329             note='Downloading playlist page',
 330             errnote='Unable to download playlist page')
 331
 332         title = self._html_search_regex(
 333             r'<a[^>]+class="parent active"[^>]*><b>\d+</b><span[^>]*>([^<]+)',
 334             webpage, 'playlist title',
 335             default=None) or self._og_search_title(webpage, fatal=False)
 336         description = self._get_user_description(webpage)
 337
 338         entries = OnDemandPagedList(
 339             functools.partial(
 340                 self._tracks_page_func,
 341                 '%s/playlists/%s' % (user_id, playlist_id), video_id, 'tracklist'),
 342             self._PAGE_SIZE)
 343
 344         return self.playlist_result(entries, video_id, title, description)
 345
 346
 347 class MixcloudStreamIE(MixcloudPlaylistBaseIE):
 348     _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/stream/?$'
 349     IE_NAME = 'mixcloud:stream'
 350
 351     _TEST = {
 352         'url': 'https://www.mixcloud.com/FirstEar/stream/',
 353         'info_dict': {
 354             'id': 'FirstEar',
 355             'title': 'First Ear',
 356             'description': 'Curators of good music\nfirstearmusic.com',
 357         },
 358         'playlist_mincount': 192,
 359     }
 360
 361     def _real_extract(self, url):
 362         user_id = self._match_id(url)
 363
 364         webpage = self._download_webpage(url, user_id)
 365
 366         entries = []
 367         prev_page_url = None
 368
 369         def _handle_page(page):
 370             entries.extend(self._find_urls_in_page(page))
 371             return self._search_regex(
 372                 r'm-next-page-url="([^"]+)"', page,
 373                 'next page URL', default=None)
 374
 375         next_page_url = _handle_page(webpage)
 376
 377         for idx in itertools.count(0):
 378             if not next_page_url or prev_page_url == next_page_url:
 379                 break
 380
 381             prev_page_url = next_page_url
 382             current_page = int(self._search_regex(
 383                 r'\?page=(\d+)', next_page_url, 'next page number'))
 384
 385             next_page_url = _handle_page(self._fetch_tracks_page(
 386                 '%s/stream' % user_id, user_id, 'stream', idx,
 387                 real_page_number=current_page))
 388
 389         username = self._og_search_title(webpage)
 390         description = self._get_user_description(webpage)
 391
 392         return self.playlist_result(entries, user_id, username, description)