X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9815bb0a551468e4939cacfffbc2d5cb8dd12431..abc36f0474141930338eef35e2444365fab932b6:/youtube_dl/extractor/mixcloud.py diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 8245b55..a24b316 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -1,115 +1,310 @@ -import json +from __future__ import unicode_literals + +import base64 +import functools +import itertools import re -import socket from .common import InfoExtractor +from ..compat import ( + compat_chr, + compat_ord, + compat_urllib_parse_unquote, + compat_urlparse, +) from ..utils import ( - compat_http_client, - compat_str, - compat_urllib_error, - compat_urllib_request, - + clean_html, ExtractorError, + OnDemandPagedList, + str_to_int, ) class MixcloudIE(InfoExtractor): - _WORKING = False # New API, but it seems good http://www.mixcloud.com/developers/documentation/ - _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([\w\d-]+)/([\w\d-]+)' - IE_NAME = u'mixcloud' - - def report_download_json(self, file_id): - """Report JSON download.""" - self.to_screen(u'Downloading json') - - def get_urls(self, jsonData, fmt, bitrate='best'): - """Get urls from 'audio_formats' section in json""" - try: - bitrate_list = jsonData[fmt] - if bitrate is None or bitrate == 'best' or bitrate not in bitrate_list: - bitrate = max(bitrate_list) # select highest - - url_list = jsonData[fmt][bitrate] - except TypeError: # we have no bitrate info. - url_list = jsonData[fmt] - return url_list - - def check_urls(self, url_list): - """Returns 1st active url from list""" - for url in url_list: - try: - compat_urllib_request.urlopen(url) - return url - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error): - url = None - - return None - - def _print_formats(self, formats): - print('Available formats:') - for fmt in formats.keys(): - for b in formats[fmt]: - try: - ext = formats[fmt][b][0] - print('%s\t%s\t[%s]' % (fmt, b, ext.split('.')[-1])) - except TypeError: # we have no bitrate info - ext = formats[fmt][0] - print('%s\t%s\t[%s]' % (fmt, '??', ext.split('.')[-1])) - break + _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': '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': r're:https?://.*\.jpg', + 'view_count': int, + }, + }, { + 'url': 'http://www.mixcloud.com/gillespeterson/caribou-7-inch-vinyl-mix-chat/', + 'info_dict': { + 'id': 'gillespeterson-caribou-7-inch-vinyl-mix-chat', + 'ext': 'mp3', + 'title': 'Caribou 7 inch Vinyl Mix & Chat', + 'description': 'md5:2b8aec6adce69f9d41724647c65875e8', + 'uploader': 'Gilles Peterson Worldwide', + 'uploader_id': 'gillespeterson', + 'thumbnail': 're:https?://.*', + 'view_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')) + + 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) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - # extract uploader & filename from url - uploader = mobj.group(1).decode('utf-8') - file_id = uploader + "-" + mobj.group(2).decode('utf-8') - - # construct API request - file_url = 'http://www.mixcloud.com/api/1/cloudcast/' + '/'.join(url.split('/')[-3:-1]) + '.json' - # retrieve .json file with links to files - request = compat_urllib_request.Request(file_url) - try: - self.report_download_json(file_url) - jsonData = compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to retrieve file: %s' % compat_str(err)) - - # parse JSON - json_data = json.loads(jsonData) - player_url = json_data['player_swf_url'] - formats = dict(json_data['audio_formats']) - - req_format = self._downloader.params.get('format', None) - - if self._downloader.params.get('listformats', None): - self._print_formats(formats) - return - - if req_format is None or req_format == 'best': - for format_param in formats.keys(): - url_list = self.get_urls(formats, format_param) - # check urls - file_url = self.check_urls(url_list) - if file_url is not None: - break # got it! - else: - if req_format not in formats: - raise ExtractorError(u'Format is not available') - - url_list = self.get_urls(formats, req_format) - file_url = self.check_urls(url_list) - format_param = req_format - - return [{ - 'id': file_id.decode('utf-8'), - 'url': file_url.decode('utf-8'), - 'uploader': uploader.decode('utf-8'), - 'upload_date': None, - 'title': json_data['name'], - 'ext': file_url.split('.')[-1].decode('utf-8'), - 'format': (format_param is None and u'NA' or format_param.decode('utf-8')), - 'thumbnail': json_data['thumbnail_url'], - 'description': json_data['description'], - 'player_url': player_url.decode('utf-8'), - }] + uploader = mobj.group(1) + cloudcast_name = mobj.group(2) + track_id = compat_urllib_parse_unquote('-'.join((uploader, cloudcast_name))) + + webpage = self._download_webpage(url, track_id) + + 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'] + + 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'([0-9,.]+)', + r'm-tooltip=["\']([\d,.]+) plays'], + webpage, 'play count', default=None)) + + return { + 'id': track_id, + 'title': title, + 'url': song_url, + 'description': description, + 'thumbnail': thumbnail, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'view_count': view_count, + } + + +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)