X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fmixcloud.py;h=202c05dcbffe082dafd338e851e78c92f279efa6;hb=96d315c2be1c00339b4cb77e80b3d5f63770a735;hp=30b33e7e98581fb0142d30837276dc476f6acd49;hpb=6d671695092236af75c49e8a74d19b771e1b4d79;p=youtube-dl diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 30b33e7e9..202c05dcb 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -1,24 +1,28 @@ from __future__ import unicode_literals +import base64 +import functools +import itertools import re from .common import InfoExtractor from ..compat import ( + compat_chr, + compat_ord, compat_urllib_parse_unquote, - compat_urllib_request + compat_urlparse, ) from ..utils import ( + clean_html, ExtractorError, - HEADRequest, - NO_DEFAULT, + OnDemandPagedList, parse_count, str_to_int, - clean_html ) 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 = [{ @@ -43,22 +47,25 @@ 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 _check_url(self, url, track_id, ext): - try: - # We only want to know if the request succeed - # don't download the whole file - self._request_webpage( - HEADRequest(url), track_id, - 'Trying %s URL' % ext) - return True - except ExtractorError: - return False + # 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) @@ -72,19 +79,15 @@ class MixcloudIE(InfoExtractor): r'(?s)]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)', webpage, 'error message', default=None) - preview_url = self._search_regex( - r'\s(?:data-preview-url|m-preview)="([^"]+)"', - webpage, 'preview url', default=None if message else NO_DEFAULT) + 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: + if message and 'stream_url' not in play_info: raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True) - song_url = re.sub(r'audiocdn(\d+)', r'stream\1', preview_url) - song_url = song_url.replace('/previews/', '/c/originals/') - if not self._check_url(song_url, track_id, 'mp3'): - song_url = song_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/') - if not self._check_url(song_url, track_id, 'm4a'): - raise ExtractorError('Unable to extract track url') + song_url = play_info['stream_url'] PREFIX = ( r'm-play-on-spacebar[^>]+' @@ -102,11 +105,11 @@ class MixcloudIE(InfoExtractor): description = self._og_search_description(webpage) like_count = parse_count(self._search_regex( r'\bbutton-favorite[^>]+>.*?]+class=["\']toggle-number[^>]+>\s*([^<]+)', - webpage, 'like count', fatal=False)) + 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, @@ -121,191 +124,199 @@ class MixcloudIE(InfoExtractor): } -class MixcloudUserIE(InfoExtractor): - """ - Information extractor for Mixcloud users. - It can retrieve a list of a user's uploads, favorites or listens. - """ +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', + 'id': 'dholbach_uploads', 'title': 'Daniel Holbach (uploads)', 'description': 'md5:327af72d1efeb404a8216c27240d1370', }, - 'playlist_mincount': 11 + 'playlist_mincount': 11, }, { 'url': 'http://www.mixcloud.com/dholbach/uploads/', 'info_dict': { - 'id': 'dholbach/uploads', + 'id': 'dholbach_uploads', 'title': 'Daniel Holbach (uploads)', 'description': 'md5:327af72d1efeb404a8216c27240d1370', }, - 'playlist_mincount': 11 + 'playlist_mincount': 11, }, { 'url': 'http://www.mixcloud.com/dholbach/favorites/', 'info_dict': { - 'id': 'dholbach/favorites', + 'id': 'dholbach_favorites', 'title': 'Daniel Holbach (favorites)', 'description': 'md5:327af72d1efeb404a8216c27240d1370', }, - 'playlist_mincount': 244 + 'params': { + 'playlist_items': '1-100', + }, + 'playlist_mincount': 100, }, { 'url': 'http://www.mixcloud.com/dholbach/listens/', 'info_dict': { - 'id': 'dholbach/listens', + 'id': 'dholbach_listens', 'title': 'Daniel Holbach (listens)', 'description': 'md5:327af72d1efeb404a8216c27240d1370', }, - 'playlist_mincount': 846 + 'params': { + 'playlist_items': '1-100', + }, + 'playlist_mincount': 100, }] - def _fetch_tracks(self, base_url, video_id, dl_note=None, dl_errnote=None): - # retrieve all fragments of a list of tracks with fake AJAX calls - track_urls = [] - current_page = 1 - while True: - # fake a AJAX request to retrieve a list fragment - page_url = base_url + "?page=%d&list=main&_ajax=1" % current_page - req = compat_urllib_request.Request(page_url, headers={"X-Requested-With": "XMLHttpRequest"}) - resp = self._download_webpage(req, video_id, note=dl_note + " (page %d)" % current_page, errnote=dl_errnote) - - # extract all track URLs from fragment - urls = re.findall(r'm-play-button m-url="(?P[^"]+)"', resp) - # clean up URLs - urls = map(clean_html, urls) - # create absolute URLs - urls = map(lambda u: "https://www.mixcloud.com" + u, urls) - track_urls.extend(urls) - - # advance to next fragment, if any - if " m-next-page-url=" in resp: - current_page += 1 - else: - break - - return track_urls - - def _handle_track_urls(self, urls): - return map(lambda u: self.url_result(u, "Mixcloud"), urls) - - def _get_user_description(self, page_content): - return self._html_search_regex( - r'
.*?

(.*?)

', - page_content, - "user description", - fatal=False) - - def _get_username(self, page_content): - return self._og_search_title(page_content) - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - user_id = mobj.group("user") - list_type = mobj.group("type") + 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" + list_type = 'uploads' - video_id = "%s/%s" % (user_id, list_type) + video_id = '%s_%s' % (user_id, list_type) - # download the user's profile to retrieve some metadata - profile = self._download_webpage("https://www.mixcloud.com/%s/" % user_id, - video_id, - note="Downloading user profile", - errnote="Unable to download user profile") + 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._get_username(profile) + username = self._og_search_title(profile) description = self._get_user_description(profile) - # retrieve all page fragments of uploads, favorites or listens - track_urls = self._fetch_tracks( - "https://www.mixcloud.com/%s/%s/" % (user_id, list_type), - video_id, - dl_note="Downloading list of %s" % list_type, - dl_errnote="Unable to download list of %s" % list_type) + 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) - # let MixcloudIE handle each track URL - entries = self._handle_track_urls(track_urls) + return self.playlist_result( + entries, video_id, '%s (%s)' % (username, list_type), description) - return { - '_type': 'playlist', - 'entries': entries, - 'title': "%s (%s)" % (username, list_type), - 'id': video_id, - "description": description - } - - -class MixcloudPlaylistIE(MixcloudUserIE): - """ - Information extractor for Mixcloud playlists. - """ +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/playlists/tokyo-finalists-2015', + 'id': 'RedBullThre3style_tokyo-finalists-2015', 'title': 'National Champions 2015', 'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3', }, - 'playlist_mincount': 16 + 'playlist_mincount': 16, }, { 'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/', 'info_dict': { - 'id': 'maxvibes/playlists/jazzcat-on-ness-radio', + 'id': 'maxvibes_jazzcat-on-ness-radio', 'title': 'Jazzcat on Ness Radio', 'description': 'md5:7bbbf0d6359a0b8cda85224be0f8f263', }, 'playlist_mincount': 23 }] - def _get_playlist_title(self, page_content): - return self._html_search_regex( - r'(?P.*?)</span>', - page_content, - "playlist title", - group="title", - fatal=True - ) - 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/playlists/%s" % (user_id, playlist_id) + user_id = mobj.group('user') + playlist_id = mobj.group('playlist') + video_id = '%s_%s' % (user_id, playlist_id) - # download the playlist page to retrieve some metadata - profile = self._download_webpage(url, - user_id, - note="Downloading playlist page", - errnote="Unable to download playlist page") + 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._get_playlist_title(profile) + playlist_title = self._html_search_regex( + r'<span[^>]+class="[^"]*list-playlist-title[^"]*"[^>]*>(.*?)</span>', + profile, 'playlist title') - # retrieve all page fragments of playlist - track_urls = self._fetch_tracks( - "https://www.mixcloud.com/%s/playlists/%s/" % (user_id, playlist_id), - video_id, - dl_note="Downloading tracklist of %s" % playlist_title, - dl_errnote="Unable to tracklist of %s" % playlist_title) + entries = OnDemandPagedList( + functools.partial( + self._tracks_page_func, + '%s/playlists/%s' % (user_id, playlist_id), video_id, 'tracklist'), + self._PAGE_SIZE) - # let MixcloudIE handle each track - entries = self._handle_track_urls(track_urls) + return self.playlist_result(entries, video_id, playlist_title, description) - return { - '_type': 'playlist', - 'entries': entries, - 'title': playlist_title, - 'id': video_id, - "description": 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)