X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fmixcloud.py;h=4ba2310fd79f09777946cff8f95d001eba12979d;hb=ec85ded83cbfa652ba94cb080aab52d8b270212a;hp=21aea0c5513ab30cbe27384351229c42e864fab3;hpb=aae53774f247edcca7671434c16f7931d5c7b9da;p=youtube-dl diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 21aea0c55..4ba2310fd 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -1,33 +1,40 @@ from __future__ import unicode_literals -import re +import base64 +import functools import itertools +import re from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, + compat_chr, + compat_ord, + compat_urllib_parse_unquote, + compat_urlparse, ) from ..utils import ( + clean_html, ExtractorError, - HEADRequest, + OnDemandPagedList, + parse_count, str_to_int, ) class MixcloudIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/([^/]+)' + _VALID_URL = r'https?://(?:(?:www|beta|m)\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)' IE_NAME = 'mixcloud' _TESTS = [{ 'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/', 'info_dict': { 'id': 'dholbach-cryptkeeper', - 'ext': 'mp3', + 'ext': 'm4a', 'title': 'Cryptkeeper', '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, }, @@ -40,47 +47,47 @@ class MixcloudIE(InfoExtractor): 'description': 'md5:2b8aec6adce69f9d41724647c65875e8', 'uploader': 'Gilles Peterson Worldwide', 'uploader_id': 'gillespeterson', - 'thumbnail': 're:https?://.*/images/', + '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, }] - def _get_url(self, track_id, template_url, server_number): - boundaries = (1, 30) - for nr in server_numbers(server_number, boundaries): - url = template_url % nr - try: - # We only want to know if the request succeed - # don't download the whole file - self._request_webpage( - HEADRequest(url), track_id, - 'Checking URL %d/%d ...' % (nr, boundaries[-1])) - return url - except ExtractorError: - pass - return None + # 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')) + + return ''.join([ + compat_chr(compat_ord(ch) ^ compat_ord(KEY[idx % len(KEY)])) + for idx, ch in enumerate(play_info)]) def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) uploader = mobj.group(1) cloudcast_name = mobj.group(2) - track_id = compat_urllib_parse.unquote('-'.join((uploader, cloudcast_name))) + track_id = compat_urllib_parse_unquote('-'.join((uploader, cloudcast_name))) webpage = self._download_webpage(url, track_id) - preview_url = self._search_regex( - r'\s(?:data-preview-url|m-preview)="([^"]+)"', webpage, 'preview url') - song_url = preview_url.replace('/previews/', '/c/originals/') - server_number = int(self._search_regex(r'stream(\d+)', song_url, 'server number')) - template_url = re.sub(r'(stream\d*)', 'stream%d', song_url) - final_song_url = self._get_url(track_id, template_url, server_number) - if final_song_url is None: - self.to_screen('Trying with m4a extension') - template_url = template_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/') - final_song_url = self._get_url(track_id, template_url, server_number) - if final_song_url is None: - raise ExtractorError('Unable to extract track url') + message = self._html_search_regex( + r'(?s)]+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[^>]+' @@ -96,18 +103,18 @@ class MixcloudIE(InfoExtractor): uploader_id = self._search_regex( r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False) description = self._og_search_description(webpage) - like_count = str_to_int(self._search_regex( - r'\bbutton-favorite\b.+m-ajax-toggle-count="([^"]+)"', - webpage, 'like count', fatal=False)) + like_count = parse_count(self._search_regex( + r'\bbutton-favorite[^>]+>.*?]+class=["\']toggle-number[^>]+>\s*([^<]+)', + webpage, 'like count', default=None)) view_count = str_to_int(self._search_regex( [r'([0-9,.]+)'], - webpage, 'play count', fatal=False)) + webpage, 'play count', default=None)) return { 'id': track_id, 'title': title, - 'url': final_song_url, + 'url': song_url, 'description': description, 'thumbnail': thumbnail, 'uploader': uploader, @@ -117,33 +124,199 @@ class MixcloudIE(InfoExtractor): } -def server_numbers(first, boundaries): - """ Server numbers to try in descending order of probable availability. - Starting from first (i.e. the number of the server hosting the preview file) - and going further and further up to the higher boundary and down to the - lower one in an alternating fashion. Namely: - - server_numbers(2, (1, 5)) - - # Where the preview server is 2, min number is 1 and max is 5. - # Yields: 2, 3, 1, 4, 5 - - Why not random numbers or increasing sequences? Since from what I've seen, - full length files seem to be hosted on servers whose number is closer to - that of the preview; to be confirmed. - """ - zip_longest = getattr(itertools, 'zip_longest', None) - if zip_longest is None: - # python 2.x - zip_longest = itertools.izip_longest - - if len(boundaries) != 2: - raise ValueError("boundaries should be a two-element tuple") - min, max = boundaries - highs = range(first + 1, max + 1) - lows = range(first - 1, min - 1, -1) - rest = filter( - None, itertools.chain.from_iterable(zip_longest(highs, lows))) - yield first - for n in rest: - yield n +class MixcloudPlaylistBaseIE(InfoExtractor): + _PAGE_SIZE = 24 + + def _find_urls_in_page(self, page): + for url in re.findall(r'm-play-button m-url="(?P[^"]+)"', 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': real_page_number, 'list': 'main', '_ajax': '1'}, + headers={'X-Requested-With': 'XMLHttpRequest'}) + + 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']+class="description-text"[^>]*>(.+?)', + page_content, 'user description', fatal=False) + + +class MixcloudUserIE(MixcloudPlaylistBaseIE): + _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P[^/]+)/(?Puploads|favorites|listens)?/?$' + IE_NAME = 'mixcloud:user' + + _TESTS = [{ + 'url': 'http://www.mixcloud.com/dholbach/', + 'info_dict': { + 'id': 'dholbach_uploads', + 'title': 'Daniel Holbach (uploads)', + 'description': 'md5:327af72d1efeb404a8216c27240d1370', + }, + 'playlist_mincount': 11, + }, { + 'url': 'http://www.mixcloud.com/dholbach/uploads/', + 'info_dict': { + 'id': 'dholbach_uploads', + 'title': 'Daniel Holbach (uploads)', + 'description': 'md5:327af72d1efeb404a8216c27240d1370', + }, + 'playlist_mincount': 11, + }, { + 'url': 'http://www.mixcloud.com/dholbach/favorites/', + 'info_dict': { + 'id': 'dholbach_favorites', + 'title': 'Daniel Holbach (favorites)', + 'description': 'md5:327af72d1efeb404a8216c27240d1370', + }, + 'params': { + 'playlist_items': '1-100', + }, + 'playlist_mincount': 100, + }, { + 'url': 'http://www.mixcloud.com/dholbach/listens/', + 'info_dict': { + 'id': 'dholbach_listens', + 'title': 'Daniel Holbach (listens)', + 'description': 'md5:327af72d1efeb404a8216c27240d1370', + }, + 'params': { + 'playlist_items': '1-100', + }, + 'playlist_mincount': 100, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user_id = mobj.group('user') + list_type = mobj.group('type') + + # if only a profile URL was supplied, default to download all uploads + if list_type is None: + list_type = 'uploads' + + video_id = '%s_%s' % (user_id, list_type) + + profile = self._download_webpage( + 'https://www.mixcloud.com/%s/' % user_id, video_id, + note='Downloading user profile', + errnote='Unable to download user profile') + + username = self._og_search_title(profile) + description = self._get_user_description(profile) + + entries = OnDemandPagedList( + functools.partial( + self._tracks_page_func, + '%s/%s' % (user_id, list_type), video_id, 'list of %s' % list_type), + self._PAGE_SIZE, use_cache=True) + + return self.playlist_result( + entries, video_id, '%s (%s)' % (username, list_type), description) + + +class MixcloudPlaylistIE(MixcloudPlaylistBaseIE): + _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P[^/]+)/playlists/(?P[^/]+)/?$' + IE_NAME = 'mixcloud:playlist' + + _TESTS = [{ + 'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/', + 'info_dict': { + 'id': 'RedBullThre3style_tokyo-finalists-2015', + 'title': 'National Champions 2015', + 'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3', + }, + '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 + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user_id = mobj.group('user') + playlist_id = mobj.group('playlist') + video_id = '%s_%s' % (user_id, playlist_id) + + profile = 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']+class="[^"]*list-playlist-title[^"]*"[^>]*>(.*?)', + profile, 'playlist title') + + entries = OnDemandPagedList( + functools.partial( + 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) + + +class MixcloudStreamIE(MixcloudPlaylistBaseIE): + _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P[^/]+)/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)