import base64
import functools
+import itertools
import re
from .common import InfoExtractor
compat_ord,
compat_urllib_parse_unquote,
compat_urlparse,
+ compat_zip
)
from ..utils import (
clean_html,
ExtractorError,
OnDemandPagedList,
- parse_count,
str_to_int,
-)
+ try_get)
class MixcloudIE(InfoExtractor):
- _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)'
+ _VALID_URL = r'https?://(?:(?:www|beta|m)\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)'
IE_NAME = 'mixcloud'
_TESTS = [{
'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
'uploader': 'Daniel Holbach',
'uploader_id': 'dholbach',
- 'thumbnail': 're:https?://.*\.jpg',
+ 'thumbnail': r're:https?://.*\.jpg',
'view_count': int,
- 'like_count': int,
},
}, {
'url': 'http://www.mixcloud.com/gillespeterson/caribou-7-inch-vinyl-mix-chat/',
'uploader_id': 'gillespeterson',
'thumbnail': 're:https?://.*',
'view_count': int,
- 'like_count': int,
},
+ }, {
+ 'url': 'https://beta.mixcloud.com/RedLightRadio/nosedrip-15-red-light-radio-01-18-2016/',
+ 'only_matching': True,
}]
- # See https://www.mixcloud.com/media/js2/www_js_2.9e23256562c080482435196ca3975ab5.js
@staticmethod
- def _decrypt_play_info(play_info):
- KEY = 'pleasedontdownloadourmusictheartistswontgetpaid'
-
- play_info = base64.b64decode(play_info.encode('ascii'))
-
+ def _decrypt_xor_cipher(key, ciphertext):
+ """Encrypt/Decrypt XOR cipher. Both ways are possible because it's XOR."""
return ''.join([
- compat_chr(compat_ord(ch) ^ compat_ord(KEY[idx % len(KEY)]))
- for idx, ch in enumerate(play_info)])
+ compat_chr(compat_ord(ch) ^ compat_ord(k))
+ for ch, k in compat_zip(ciphertext, itertools.cycle(key))])
+
+ @staticmethod
+ def _decrypt_and_extend(stream_info, url_key, getter, key, formats):
+ maybe_url = stream_info.get(url_key)
+ if maybe_url is not None:
+ decrypted = MixcloudIE._decrypt_xor_cipher(key, base64.b64decode(maybe_url))
+ formats.extend(getter(decrypted))
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
webpage = self._download_webpage(url, track_id)
+ # Legacy path
+ encrypted_play_info = self._search_regex(
+ r'm-play-info="([^"]+)"', webpage, 'play info', default=None)
+
+ if encrypted_play_info is not None:
+ # Decode
+ encrypted_play_info = base64.b64decode(encrypted_play_info)
+ else:
+ # New path
+ full_info_json = self._parse_json(self._html_search_regex(
+ r'<script id="relay-data" type="text/x-mixcloud">([^<]+)</script>', webpage, 'play info'), 'play info')
+ for item in full_info_json:
+ item_data = try_get(item, lambda x: x['cloudcast']['data']['cloudcastLookup'])
+ if try_get(item_data, lambda x: x['streamInfo']['url']):
+ info_json = item_data
+ break
+ else:
+ raise ExtractorError('Failed to extract matching stream info')
+
message = self._html_search_regex(
r'(?s)<div[^>]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)',
webpage, 'error message', default=None)
- encrypted_play_info = self._search_regex(
- r'm-play-info="([^"]+)"', webpage, 'play info')
- play_info = self._parse_json(
- self._decrypt_play_info(encrypted_play_info), track_id)
-
- if message and 'stream_url' not in play_info:
- raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
-
- song_url = play_info['stream_url']
-
- PREFIX = (
- r'm-play-on-spacebar[^>]+'
- r'(?:\s+[a-zA-Z0-9-]+(?:="[^"]+")?)*?\s+')
- title = self._html_search_regex(
- PREFIX + r'm-title="([^"]+)"', webpage, 'title')
- thumbnail = self._proto_relative_url(self._html_search_regex(
- PREFIX + r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail',
- fatal=False))
- uploader = self._html_search_regex(
- PREFIX + r'm-owner-name="([^"]+)"',
- webpage, 'uploader', fatal=False)
- uploader_id = self._search_regex(
- r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
- description = self._og_search_description(webpage)
- like_count = parse_count(self._search_regex(
- r'\bbutton-favorite[^>]+>.*?<span[^>]+class=["\']toggle-number[^>]+>\s*([^<]+)',
- webpage, 'like count', fatal=False))
- view_count = str_to_int(self._search_regex(
- [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
- r'/listeners/?">([0-9,.]+)</a>'],
- webpage, 'play count', fatal=False))
+ js_url = self._search_regex(
+ r'<script[^>]+\bsrc=["\"](https://(?:www\.)?mixcloud\.com/media/js2/www_js_4\.[^>]+\.js)',
+ webpage, 'js url', default=None)
+ if js_url is None:
+ js_url = self._search_regex(
+ r'<script[^>]+\bsrc=["\"](https://(?:www\.)?mixcloud\.com/media/js/www\.[^>]+\.js)',
+ webpage, 'js url')
+ js = self._download_webpage(js_url, track_id)
+ # Known plaintext attack
+ if encrypted_play_info:
+ kps = ['{"stream_url":']
+ kpa_target = encrypted_play_info
+ else:
+ kps = ['https://', 'http://']
+ kpa_target = base64.b64decode(info_json['streamInfo']['url'])
+ for kp in kps:
+ partial_key = self._decrypt_xor_cipher(kpa_target, kp)
+ for quote in ["'", '"']:
+ key = self._search_regex(r'{0}({1}[^{0}]*){0}'.format(quote, re.escape(partial_key)), js,
+ "encryption key", default=None)
+ if key is not None:
+ break
+ else:
+ continue
+ break
+ else:
+ raise ExtractorError('Failed to extract encryption key')
+
+ if encrypted_play_info is not None:
+ play_info = self._parse_json(self._decrypt_xor_cipher(key, encrypted_play_info), 'play info')
+ if message and 'stream_url' not in play_info:
+ raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
+ song_url = play_info['stream_url']
+ formats = [{
+ 'format_id': 'normal',
+ 'url': song_url
+ }]
+
+ title = self._html_search_regex(r'm-title="([^"]+)"', webpage, 'title')
+ thumbnail = self._proto_relative_url(self._html_search_regex(
+ r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail', fatal=False))
+ uploader = self._html_search_regex(
+ r'm-owner-name="([^"]+)"', webpage, 'uploader', fatal=False)
+ uploader_id = self._search_regex(
+ r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
+ description = self._og_search_description(webpage)
+ view_count = str_to_int(self._search_regex(
+ [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
+ r'/listeners/?">([0-9,.]+)</a>',
+ r'(?:m|data)-tooltip=["\']([\d,.]+) plays'],
+ webpage, 'play count', default=None))
+
+ else:
+ title = info_json['name']
+ thumbnail = try_get(info_json,
+ lambda x: 'https://thumbnailer.mixcloud.com/unsafe/600x600/' + x['picture']['urlRoot'])
+ uploader = try_get(info_json, lambda x: x['owner']['displayName'])
+ uploader_id = try_get(info_json, lambda x: x['owner']['username'])
+ description = try_get(info_json, lambda x: x['description'])
+ view_count = try_get(info_json, lambda x: x['plays'])
+
+ stream_info = info_json['streamInfo']
+ formats = []
+ self._decrypt_and_extend(stream_info, 'url', lambda x: [{
+ 'format_id': 'normal',
+ 'url': x
+ }], key, formats)
+ self._decrypt_and_extend(stream_info, 'hlsUrl', lambda x: self._extract_m3u8_formats(x, title), key,
+ formats)
+ self._decrypt_and_extend(stream_info, 'dashUrl', lambda x: self._extract_mpd_formats(x, title), key,
+ formats)
return {
'id': track_id,
'title': title,
- 'url': song_url,
+ 'formats': formats,
'description': description,
'thumbnail': thumbnail,
'uploader': uploader,
'uploader_id': uploader_id,
'view_count': view_count,
- 'like_count': like_count,
}
class MixcloudPlaylistBaseIE(InfoExtractor):
_PAGE_SIZE = 24
- def _fetch_tracks_page(self, path, video_id, page_name, current_page):
- resp = self._download_webpage(
+ def _find_urls_in_page(self, page):
+ for url in re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', page):
+ yield self.url_result(
+ compat_urlparse.urljoin('https://www.mixcloud.com', clean_html(url)),
+ MixcloudIE.ie_key())
+
+ def _fetch_tracks_page(self, path, video_id, page_name, current_page, real_page_number=None):
+ real_page_number = real_page_number or current_page + 1
+ return self._download_webpage(
'https://www.mixcloud.com/%s/' % path, video_id,
note='Download %s (page %d)' % (page_name, current_page + 1),
errnote='Unable to download %s' % page_name,
- query={'page': (current_page + 1), 'list': 'main', '_ajax': '1'},
+ query={'page': real_page_number, 'list': 'main', '_ajax': '1'},
headers={'X-Requested-With': 'XMLHttpRequest'})
- for url in re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', resp):
- yield self.url_result(
- compat_urlparse.urljoin('https://www.mixcloud.com', clean_html(url)),
- MixcloudIE.ie_key())
+ def _tracks_page_func(self, page, video_id, page_name, current_page):
+ resp = self._fetch_tracks_page(page, video_id, page_name, current_page)
+
+ for item in self._find_urls_in_page(resp):
+ yield item
def _get_user_description(self, page_content):
return self._html_search_regex(
- r'<div[^>]+class="description-text"[^>]*>(.+?)</div>',
+ r'<div[^>]+class="profile-bio"[^>]*>(.+?)</div>',
page_content, 'user description', fatal=False)
class MixcloudUserIE(MixcloudPlaylistBaseIE):
- _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
+ _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
IE_NAME = 'mixcloud:user'
_TESTS = [{
'info_dict': {
'id': 'dholbach_uploads',
'title': 'Daniel Holbach (uploads)',
- 'description': 'md5:327af72d1efeb404a8216c27240d1370',
+ 'description': 'md5:def36060ac8747b3aabca54924897e47',
},
'playlist_mincount': 11,
}, {
'info_dict': {
'id': 'dholbach_uploads',
'title': 'Daniel Holbach (uploads)',
- 'description': 'md5:327af72d1efeb404a8216c27240d1370',
+ 'description': 'md5:def36060ac8747b3aabca54924897e47',
},
'playlist_mincount': 11,
}, {
'info_dict': {
'id': 'dholbach_favorites',
'title': 'Daniel Holbach (favorites)',
- 'description': 'md5:327af72d1efeb404a8216c27240d1370',
+ 'description': 'md5:def36060ac8747b3aabca54924897e47',
},
'params': {
'playlist_items': '1-100',
'info_dict': {
'id': 'dholbach_listens',
'title': 'Daniel Holbach (listens)',
- 'description': 'md5:327af72d1efeb404a8216c27240d1370',
+ 'description': 'md5:def36060ac8747b3aabca54924897e47',
},
'params': {
'playlist_items': '1-100',
entries = OnDemandPagedList(
functools.partial(
- self._fetch_tracks_page,
+ self._tracks_page_func,
'%s/%s' % (user_id, list_type), video_id, 'list of %s' % list_type),
self._PAGE_SIZE, use_cache=True)
class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
- _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
+ _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
IE_NAME = 'mixcloud:playlist'
_TESTS = [{
'playlist_mincount': 16,
}, {
'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
- 'info_dict': {
- 'id': 'maxvibes_jazzcat-on-ness-radio',
- 'title': 'Jazzcat on Ness Radio',
- 'description': 'md5:7bbbf0d6359a0b8cda85224be0f8f263',
- },
- 'playlist_mincount': 23
+ 'only_matching': True,
}]
def _real_extract(self, url):
playlist_id = mobj.group('playlist')
video_id = '%s_%s' % (user_id, playlist_id)
- profile = self._download_webpage(
+ webpage = self._download_webpage(
url, user_id,
note='Downloading playlist page',
errnote='Unable to download playlist page')
- description = self._get_user_description(profile)
- playlist_title = self._html_search_regex(
- r'<span[^>]+class="[^"]*list-playlist-title[^"]*"[^>]*>(.*?)</span>',
- profile, 'playlist title')
+ title = self._html_search_regex(
+ r'<a[^>]+class="parent active"[^>]*><b>\d+</b><span[^>]*>([^<]+)',
+ webpage, 'playlist title',
+ default=None) or self._og_search_title(webpage, fatal=False)
+ description = self._get_user_description(webpage)
entries = OnDemandPagedList(
functools.partial(
- self._fetch_tracks_page,
+ self._tracks_page_func,
'%s/playlists/%s' % (user_id, playlist_id), video_id, 'tracklist'),
self._PAGE_SIZE)
- return self.playlist_result(entries, video_id, playlist_title, description)
+ return self.playlist_result(entries, video_id, title, description)
+
+
+class MixcloudStreamIE(MixcloudPlaylistBaseIE):
+ _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/stream/?$'
+ IE_NAME = 'mixcloud:stream'
+
+ _TEST = {
+ 'url': 'https://www.mixcloud.com/FirstEar/stream/',
+ 'info_dict': {
+ 'id': 'FirstEar',
+ 'title': 'First Ear',
+ 'description': 'Curators of good music\nfirstearmusic.com',
+ },
+ 'playlist_mincount': 192,
+ }
+
+ def _real_extract(self, url):
+ user_id = self._match_id(url)
+
+ webpage = self._download_webpage(url, user_id)
+
+ entries = []
+ prev_page_url = None
+
+ def _handle_page(page):
+ entries.extend(self._find_urls_in_page(page))
+ return self._search_regex(
+ r'm-next-page-url="([^"]+)"', page,
+ 'next page URL', default=None)
+
+ next_page_url = _handle_page(webpage)
+
+ for idx in itertools.count(0):
+ if not next_page_url or prev_page_url == next_page_url:
+ break
+
+ prev_page_url = next_page_url
+ current_page = int(self._search_regex(
+ r'\?page=(\d+)', next_page_url, 'next page number'))
+
+ next_page_url = _handle_page(self._fetch_tracks_page(
+ '%s/stream' % user_id, user_id, 'stream', idx,
+ real_page_number=current_page))
+
+ username = self._og_search_title(webpage)
+ description = self._get_user_description(webpage)
+
+ return self.playlist_result(entries, user_id, username, description)